The University of Southern Mississippi
The Aquila Digital Community
Master's Theses
Fall 12-2016
CAVITY RINGDOWN SPECTROSCOPY IN
NITROGEN/OXYGEN MIXTURES IN THE
PRESENCE OF ALPHA RADIATION
Sidney John Gautrau
University of Southern Mississippi
Follow this and additional works at: http://aquila.usm.edu/masters_theses
Part of the Atomic, Molecular and Optical Physics Commons, and the Optics Commons
Recommended Citation
Gautrau, Sidney John, "CAVITY RINGDOWN SPECTROSCOPY IN NITROGEN/OXYGEN MIXTURES IN THE PRESENCE
OF ALPHA RADIATION" (2016). Master's Theses. 258.
http://aquila.usm.edu/masters_theses/258
This Masters Thesis is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Master's Theses
by an authorized administrator of The Aquila Digital Community. For more information, please contact [email protected].
CAVITY RINGDOWN SPECTROSCOPY IN NITROGEN/OXYGEN MIXTURES IN
THE PRESENCE OF ALPHA RADIATION
by
Sidney John Gautrau
A Thesis
Submitted to the Graduate School
and the Department of Physics and Astronomy
at The University of Southern Mississippi
in Partial Fulfillment of the Requirements
for the Degree of Master of Science
Approved:
________________________________________________
Dr. Christopher B. Winstead, Committee Chair
Professor, Physics and Astronomy
________________________________________________
Dr. Ras B. Pandey, Committee Member
Professor, Physics and Astronomy
________________________________________________
Dr. Michael D. Vera, Committee Member
Associate Professor, Physics and Astronomy
________________________________________________
Dr. Karen S. Coats
Dean of the Graduate School
December 2016
COPYRIGHT BY
Sidney John Gautrau
2016
Published by the Graduate School
ABSTRACT
CAVITY RINGDOWN SPECTROSCOPY IN NITROGEN/OXYGEN MIXTURES IN
THE PRESENCE OF ALPHA RADIATION
by Sidney John Gautrau
December 2016
This research was part of an effort to experimentally validate computational
models under development for radiation-induced atmospheric effects. Cavity Ringdown
Spectroscopy (CRDS) was used to measure the concentration of chemical products
generated as a result of radiation interactions in a controlled atmosphere. Experiments
were conducted in a vacuum chamber interfaced with a gas introduction system that
controlled the initial atmospheric composition. A quadrupole mass spectrometer and
tunable dye laser were integrated to confirm initial atmospheric composition, and provide
wavelength flexibility for detecting a variety of chemical products generated by radiation
interactions. CRDS measurements were made for ozone production resulting from alpha
radiation interactions in nitrogen-oxygen mixtures while the Polonium 210 alpha sources
were cycled between being shielded and exposed.
The beginning of this thesis will provide brief reviews of ionizing radiation and
ozone formation, along with explanations of Cavity Enhanced Absorption Spectroscopy
and CRDS. This is followed by a description of the existing vacuum and optical system,
as well as the modifications required to integrate the mass spectrometer and dye laser.
Another chapter details the modifications made to the existing software, as well as new
software developed for the mass spectrometer and dye laser. Finally, results are
presented showing the rapid production of ozone following exposure of the alpha
ii
radiation to the controlled atmosphere. Results also show ozone concentrations rapidly
decreasing when the source was re-shielded, but when the source was re-exposed, the
ozone concentration started at the previous concentration and continued to increase. This
indicates the presence of a possible ozone catalyst.
iii
ACKNOWLEDGMENTS
I would like to thank research advisor Dr. Chris Winstead for providing a project
and environment that promoted creativity and curiosity. I would also like to thank the
rest of the committee, Dr. Ras Pandey and Dr. Michael Vera, for all of their advice and
time throughout this process. I would like to thank lab partners Tyler Reese and Patrick
Ables. Tyler helped bring me up to speed on the existing system and helped me when I
had any problem with any equipment. I will reference Tyler’s thesis throughout this
thesis and would like to reference Tyler’s thesis here because I used an outline similar to
Tyler’s 1. Patrick and I spent countless hours collaborating and working on the software
to make the most user-friendly system for future students with the equipment that was
available to us. Patrick built the foundation on which all the software was later modified
or written.
Tyler Reese, “A System for Measuring Radiation Induced Chemical Products in Atmospheric Gases
Using Optical Detection Methods” (Master’s Thesis, The University of Southern Mississippi, 2010), v.
1
iv
DEDICATION
I would like to dedicate this thesis to my mother and late father.
v
TABLE OF CONTENTS
ABSTRACT ........................................................................................................................ ii
ACKNOWLEDGMENTS ................................................................................................. iv
DEDICATION .................................................................................................................... v
LIST OF TABLES ............................................................................................................. ix
LIST OF ILLUSTRATIONS .............................................................................................. x
LIST OF ABBREVIATIONS .......................................................................................... xiii
CHAPTER I – INTRODUCTION ...................................................................................... 1
CHAPTER II – BRIEF REVIEW OF RADIOACTIVE DECAY...................................... 3
CHAPTER III – OZONE PRODUCTION ......................................................................... 8
Ozone Formation in the Atmosphere .............................................................................. 8
Ozone Formation from Ionizing Radiation ................................................................... 12
CHAPTER IV – CAVITY ENHANCED ABSORPTION SPECTROSCOPY ............... 15
CHAPTER V – EXISTING SYSTEMS TO BE UTILIZED ........................................... 23
Existing Vacuum System .............................................................................................. 23
Physical Characteristics ............................................................................................ 24
Pumping Mechanisms ............................................................................................... 27
Chamber Measurement Capabilities ......................................................................... 27
System Control.......................................................................................................... 28
Existing CRDS System ................................................................................................. 29
vi
CHAPTER VI – SYSTEM MODIFICATIONS ............................................................... 32
Laser System ................................................................................................................. 32
Universal Gas Analyzer (UGA) .................................................................................... 34
Measurement Device .................................................................................................... 37
CHAPTER VII – SOFTWARE DEVELOPMENT .......................................................... 40
Laboratory Network ...................................................................................................... 41
Windows XP Computer for Vacuum System, Dye Laser, and Mass Spectrometer . 43
“Real Time Panel.vi.” ........................................................................................... 43
“Alarm Conditions.vi” .......................................................................................... 43
“LCD Display Loop.vi” ........................................................................................ 43
“Push Button Panel.vi, ControlsSubVI.vi, Status Text.vi” ................................... 44
“Vacuum System Contol.vi, Cycle Control.vi, and Rad Cover Cycle.vi”............ 45
“Dye Laser Control.vi” ......................................................................................... 51
“UGA.vi” .............................................................................................................. 54
Windows XP Computer for Data Collection ............................................................ 56
“PSP Variables for LV9XP.vi” ............................................................................. 57
“Ringdown Time.vi” ............................................................................................. 58
Experiment Automation .................................................................................... 62
Windows 7 Computer for Remote Login and Email Notifications .......................... 63
“Alarm for Win7.vi” ............................................................................................. 63
vii
CHAPTER VIII – DATA COLLECTION AND RESULTS ........................................... 65
CHAPTER IX – CONCLUSIONS AND FUTURE WORK ............................................ 86
APPENDIX A – Permission Emails ................................................................................. 93
REFERENCES ................................................................................................................. 96
viii
LIST OF TABLES
Table 1 Penetration Depth Comparison for Representative Radiation Sources ................. 7
Table 2 Mirror Configurations .......................................................................................... 34
Table 3 PMT Filter Configurations................................................................................... 39
Table 4 Predicted Shift of Average Ringdown Values from Vacuum-N2 Cycles for
Rayleigh Scattering ........................................................................................................... 66
Table 5 Predicted Number Density from Vacuum-N2 Cycles for Rayleigh Scattering .... 68
Table 6 Radiation Source Manufactured-Date and Date Experiments Were Performed . 84
Table 7 Predicted Concentrations of Other Potential Products of Ionizing Radiation
Resulting from the Observed Shift in Ringdown Time .................................................... 86
Table 8 Predicted Concentrations of Products of Ionizing Radiation Resulting from the
Observed Drift in Ringdown Time of τ0 Measurements................................................... 88
ix
LIST OF ILLUSTRATIONS
Figure 1. Ozone Absorption Spectrum. ............................................................................ 16
Figure 2. Basic representation of Beer’s Law. ................................................................. 17
Figure 3. Basic representation of CEAS........................................................................... 18
Figure 4. Vacuum chamber port use and location diagram. ............................................. 24
Figure 5. Previously developed vacuum system. ............................................................. 25
Figure 6. Aluminum source holder and radiation source cover. ...................................... 26
Figure 7. Polonium 210 Radiation Source ....................................................................... 27
Figure 8. Old cRIO. .......................................................................................................... 28
Figure 9. Previously modified 8” flange for CRD cavity. ................................................ 29
Figure 10. Previously used laser configuration. ............................................................... 30
Figure 11. PXI used for data acquisition. ......................................................................... 31
Figure 12. Dye Laser and Nd: YAG laser. ....................................................................... 32
Figure 13. Small ND: YAG in use. .................................................................................. 33
Figure 14. UGA table with adjustable height. .................................................................. 35
Figure 15. Swagelok stainless steel Ultra-Torr fitting used for UGA capillary. .............. 36
Figure 16. UGA mass spectrum ....................................................................................... 37
Figure 17. PMT used in 266nm configuration. ................................................................ 38
Figure 18. Laboratory network diagram. .......................................................................... 42
Figure 19. Vacuum cart push panel. ................................................................................. 44
Figure 20. Front panel of Vacuum System Control VI on “Controls” tab. ...................... 46
Figure 21. Front panel of Vacuum System Control VI on “Automated Task” tab. ......... 48
Figure 22. Front panel of Dye Laser Control VI on “Lock” tab. ..................................... 52
x
Figure 23. Front panel of Dye Laser Control VI on “Sirah Control” tab. ........................ 54
Figure 24. Front panel of UGA VI. .................................................................................. 56
Figure 25. Front panel of Ringdown VI and PSP Variables VI. ...................................... 57
Figure 26. Front Panel of Ringdown Time VI. ................................................................ 58
Figure 27. Front panel of Ringdown VI in Align Mode................................................... 59
Figure 28. Front panel of Alarm VI.................................................................................. 64
Figure 29. Ringdown times (ns) vs time (hrs) of N2 pressure cycles at 266 nm. ............. 67
Figure 30. SO2 absorption cross section from 238 nm – 325 nm. .................................... 69
Figure 31. Experimental SO2 absorption cross section compared to literature reports. ... 71
Figure 32. Ozone concentrations after one-hour τ0, five ten-minute radiation cycles with
20% O2 in N2..................................................................................................................... 73
Figure 33. Ringdown times of one-hour τ0, five ten-minute radiation cycles with 20% O2
in N2. ................................................................................................................................. 75
Figure 34. Ringdown times of two-hour τ0, five ten-minute radiation cycles with 20% O2
in N2. ................................................................................................................................. 77
Figure 35. Ringdown times of three-hour τ0, five ten-minute radiation cycles with 20%
O2 in N2. ............................................................................................................................ 78
Figure 36. Ozone concentration from one, two, and three-hour τ0 measurements as a
function of time from when the chamber was filled. ........................................................ 79
Figure 37. Ringdown times of two-hour τ0, five one-hour radiation cycles with 20% O2 in
N2. ..................................................................................................................................... 80
Figure 38. Ozone concentrations after two-hour τ0, five one-hour radiation cycles with
20% O2 in N2..................................................................................................................... 81
xi
Figure 39. Ozone concentrations of ten-minute and one-hour radiation cover cycle
following three hours of shielding with 20% O2 in N2. .................................................... 82
Figure 40. Ozone concentrations after one-hour τ0, five ten-minute radiation cycles with
zero, ten, and twenty percent oxygen balanced in nitrogen. ............................................. 83
Figure 41. Early development of Cantera results showing the production and destruction
of ozone during shielded/exposed cycles. ......................................................................... 90
Figure 42. Enlarged view of the ozone data from Figure 41. ........................................... 90
Figure 36. Ozone concentration from one, two, and three-hour τ0 measurements as a
function of time from when the chamber was filled. ........................................................ 91
xii
LIST OF ABBREVIATIONS
CEAS
Cavity Enhanced Absorption Spectroscopy
CRD
Cavity Ringdown
CRDS
Cavity Ringdown Spectroscopy
EM
Electromagnetic
GUI
Graphic User Interface
ND: YAG
Neodymiun-doped Yttrium Aluminum
NI
National Instruments
NI-PSP
NI Publish-Subscribe Protocol
PMT
Photomultiplier Tube
ppb
parts per billion
ppm
parts per million
SVE
LabVIEW’s Shared Variable Engine
SRS
Stanford Research Systems
UGA
Universal Gas Analyzer
UV
Ultra-violet
VI
LabVIEW’s Virtual Instrument
Garnet
xiii
CHAPTER I – INTRODUCTION
Currently, the most common way to measure radiation levels is by direct
incidence of the radiation on a detection device. Most detection devices put the operator
of the device at risk of being exposed to potentially harmful radiation. This research was
performed on a funded grant and is part of an effort to create a computational model for
the effects of radiation interactions under different atmospheric conditions. The
experimental work performed in this thesis will be used toward validating a
computational model for radiation-induced atmospheric chemistry that is under
development within our research group. With the use of a working model, unique
chemical products may be discovered that could lead to an alternative method of
detecting radiation from a safer distance.
Emitted particles from radioactive decay interact with the atoms and molecules in
the air around the radiation source and initiate various chemical reactions which can
generate a variety of chemical products. The chemical products and the rates at which
the chemical products are created depends directly on the atmospheric composition and
the radiation source present. The goal of this research was to detect those chemical
products using Cavity Enhanced Absorption Spectroscopy (CEAS) inside an existing
vacuum chamber equipped with a Cavity Ringdown (CRD) cavity. A quadrupole mass
spectrometer and tunable dye laser were also integrated into the CRD system to confirm
the initial atmospheric composition and provide wavelength flexibility for detecting a
variety of chemical products by future students. For this research, ozone was measured
for various atmospheric conditions and radioactive activities.
1
This thesis will include brief reviews of ionizing radiation and ozone formation,
explanations of CEAS and Cavity Ringdown Spectroscopy (CRDS), descriptions of the
existing vacuum and optical systems, the modifications required to integrate the mass
spectrometer and dye laser, and the purpose of every program used by this system with
brief details for future students. Experiments were conducted for calibration of the
CRDS system, and experiments were performed where Polonium 210 alpha sources were
cycled multiple times between a shielded and an exposed status for different time periods.
Results will be presented showing the rapid production of ozone following the exposure
of the source. When the source was re-shielded, the ozone concentrations rapidly
decreased to nearly zero. For each successive cycle, the ozone concentration started at
the previous cycle’s final concentration and continued to increase. In between these
increasing cycles, the ozone concentration reacted away to zero. This suggests that the
radiation caused changes to the atmosphere that aided in the formation of ozone, but
these changes were also stable over the time period that the source was shielded.
Preliminary trials of the chemical kinetics software have shown results with similar
behavior and will be discussed. The efforts of this project were intended to extend the
capabilities of the system to allow for CRDS measurements at different wavelengths in
search of different chemical products produced by radiation interactions.
2
CHAPTER II – BRIEF REVIEW OF RADIOACTIVE DECAY
Radioactivity is the process of the spontaneous decay and transformation of an
unstable atomic nucleus by the emission of a high-energy nuclear particle and/or
electromagnetic (EM) radiation 2. After a particle or EM wave is emitted from an
unstable nucleus, the lower energy nucleus either contains a different number of protons
and neutrons or the nucleus is in a different state. A stable nucleus in its lowest energy
state has no reason to spontaneously decay. In some cases, a radioactive nucleus can
decay into another unstable nucleus which can undergo another decay and repeat this
process until a stable nucleus is produced. The subsequent nuclei are known as daughter
nuclei. Some nuclei are already stable, and are therefore, not radioactive. Some known
radioactive elements have isotopes, nuclei of the same element with different numbers of
neutrons, that are not radioactive 3.
The exact time at which a decay will occur cannot be predicted because
radioactive decay is spontaneous. Radioactive decays do occur consistently enough that
over a sufficiently long time period, an average decay rate can be observed. The
Becquerel, abbreviated Bq, is a unit of radioactivity that is equal to one decay per second
(s-1). The Curie, abbreviated Ci, is equal to 3.7 x 1010 decays per second or 37 GBq.
The rate of decay of an isotope is directly related to the number of atoms present
in the following expression:
− 𝑑𝑁⁄𝑑𝑡 = 𝜆𝑁
(2.1)
Michael F. L’Annunziata, Radioactivity: Introduction and History (Elsevier, 2007), 1.
Dudley G. Miller, Radioactivity and Radiation Detection (Gordon and Breach Science Publishers, 1972),
chap. 1.
2
3
3
where 𝑁 is the number of radioactive atoms present, 𝑡 is time, and 𝜆 is the decay rate
constant. After integration, the expression becomes the following:
𝑁 = 𝑁0 𝑒 −λt
(2.2)
where 𝑁0 is the beginning number of radioactive atoms present, and 𝑁 is the number of
radioactive atoms remaining after time 𝑡. By rearranging the equation, the half life can
be obtained. The half life is the time it takes for a quantity of atoms to decay to half the
initial amount.
1⁄ = 𝑁⁄ = exp (−𝜆𝑡1 )
⁄2
2
𝑁0
𝑡1⁄ = 𝑙𝑛2⁄𝜆
2
(2.3)
where 𝑡1⁄ is the half life time, and 𝜆 is the decay rate constant 4. Because the level of
2
radioactivity is directly proportional to the amount of present material, the half life is also
the time it takes for the activity level due to a particular isotope to reduce by a factor of
one half. To calculate the current activity of a sample, one only needs the initial activity
level with the date that measurement was made, and the half life.
Three major types of radioactive decays include alpha particle (α), beta particle
(β), and gamma ray (γ) emissions. These three types of emission are all forms of ionizing
radiation. These emissions carry enough energy to interact with the electrons in the
surrounding materials and are capable of ionizing the material by freeing electrons. Each
type of emission interacts with surrounding materials with different efficiencies 5.
4
J. E. Turner, Atoms, Radiation, and Radiation Protection, 3rd completely rev. and ed, Physics Textbook
(Weinheim: Wiley-VCH, 2007), chap. 4; Miller, Radioactivity and Radiation Detection, chap. 1.
5
Nicholas Tsoulfanidis, Measurement and Detection of Radiation (Hemisphere Pub. Corp., 1983), 1.
4
An alpha particle is equivalent to a helium nucleus and consists of a combination
of two protons and two neutrons. An alpha particle emission reduces the nucleus mass by
nearly four atomic mass units (amu) 6 and leaves the nucleus with a remaining charge of
Z-2, where Z is the number of protons in the nucleus. Alpha particle emission usually
occurs in the heavier elements; only a small number of elements below 83 amu are alpha
emitters 7. All alpha particles emitted by a particular element have nearly the same
energy. Relative to beta particles, alpha particles travel at a much slower velocity due to
the alpha particles’ larger mass, which allows them to be much less susceptible to
scattering and changes in direction 8. Alpha particles dissipate their energy through
collision, and can be stopped by dry air within a distance of several centimeters 9.
A beta particle is an electron. A beta particle is created when a neutron decays
into an electron and a proton; the proton remains in the nucleus and the electron is
emitted as the beta particle. A beta particle emission reduces the nucleus mass by the
mass of the emitted electron plus its binding energy and leaves the nucleus with a
remaining charge of Z+1. During a beta decay, an anti-neutrino is emitted. Even with
the anti-neutrino’s relatively small mass, the anti-neutrino is able to carry away
significant amounts of energy via kinetic energy. Unlike alpha particles, beta particles’
energies can fall into a broad band of energies up to a maximum value 10. Because of the
beta particle’s small mass, it has to travel at high velocities to carry away any significant
6
S. C. Lind, The Chemical Effects of Alpha Particles, American Chemical Society Monograph Series (New
York: The Chemical Catalog Company, 1921), chap. 2.
7
Turner, Atoms, Radiation, and Radiation Protection, 96.
8
Lind, The Chemical Effects of Alpha Particles, chap. 2.
9
S. C. Lind and D. C. Bardwell, “Ozonization and Interaction of Oxygen with Nitrogen under Alpha
Radiation,” Journal of the American Chemical Society 51, no. 9 (1929): 2751–2758; L’Annunziata,
Radioactivity, chap. 1.
10
Turner, Atoms, Radiation, and Radiation Protection, chap. 3.
5
energy. Beta particles dissipate their energy through collision, and because of the
electron’s small mass, the paths are often affected by multiple scatterings. Beta particles
can be stopped by dry air within a distance of several meters 11.
A gamma ray is a high energy photon usually resulting from an excited daughter
nuclide from an alpha or beta emission. A gamma ray does not change the mass or
charge of the nucleus because the photon is massless and has no charge. Like alpha
particles, gamma rays emitted by a particular element have nearly the same energies.
Gamma ray photons dissipate their energy through Compton scattering, photo ionization
(photoelectric effect), and particle/antiparticle pair production 12.
The type of source for this project was selected using range calculations
performed in a previous project to create a system for measuring radiation-induced
chemical products using Cavity Ringdown Spectroscopy. Table 1 shows the expected
range for a Cobalt-60, Strontium-90, and Polonium-210 source using range calculations
for dry air near sea level and was reproduced with permission (See Appendix A) 13.
William James Price, Nuclear Radiation Detection (McGraw-Hill, 1964), chap. 1; L’Annunziata,
Radioactivity, chap. 2.
12
Price, Nuclear Radiation Detection, chap. 1.
13
Reese, “A System for Measuring Radiation Induced Chemical Products in Atmospheric Gases Using
Optical Detection Methods,” 11.
11
6
Table 1
Penetration Depth Comparison for Representative Radiation Sources
Radiation
Type
Alpha
Beta
Gamma
Source
210
Po
Sr
60
Co
90
Energy Values
of Emitted
Radiation
(MeV)
5.3
0.546, 2.28
1.17, 1.33
Nearest Energy
Value with
Mass
Attenuation
Provided
(MeV)
5.5
0.55, 2.5
1.25
Expected
Range (m)
0.042
1.89, 11.4
146.2*
*Gamma range corresponds to distance at which the intensity is expected to be reduced by a factor of 1/e
The gamma emitting 60Co undergoes a beta decay into Nickel-60 which
undergoes a coincident gamma-gamma decay producing two gamma rays. A 60Co
gamma-ray has an expected range of 146.2 m. The beta particles from 90Sr have an
expected range of 1.89 m and 11.4 m from 90Sr undergoing beta decay into Yttrium-90,
which almost immediately undergoes a second beta decay into Zirconium-90. The alpha
emitting 210Po has an expected range of 0.042 m. The available apparatus containing the
components to expose a radiation source to a controlled atmosphere uses an 18” (45.72
cm) diameter spherical stainless steel chamber. The selected source must transfer all of
its emitted energy into the surrounding gas before reaching the chamber wall.
210
Po was
selected because of its small expected range. The 210Po source also deposits more energy
per volume than a beta or gamma source of similar activity would.
7
CHAPTER III – OZONE PRODUCTION
Ozone production is dominated by the single oxygen atom dependent reaction
𝑂 + 𝑂2 + 𝑀 → 𝑂3 + 𝑀
(3.1)
where molecular oxygen (𝑂2) and atomic oxygen (𝑂) react with a third body collision
partner (𝑀) to produce ozone (𝑂3) 14. The third body is required to carry away the excess
energy from the reaction, and for controlled atmospheres consisting of only molecular
nitrogen (N2) and oxygen (O2), the third body will most often be N2, O2, or O3 15.
The main reaction responsible for producing free oxygen atoms is different in the
upper atmosphere than for a radiation source in a controlled atmospheric chamber. Free
oxygen atoms are formed in the upper atmosphere when sunlight is present, and free
oxygen atoms can be formed in a controlled atmospheric simulation chamber when
ionizing radiation is present. The production and destruction of ozone in the atmosphere
will be discussed because of possible similarities of ozone destruction when the radiation
source was shielded after being exposed in this thesis. Then, ozone formation from
ionizing radiation will be discussed.
Ozone Formation in the Atmosphere
Ozone formation has been studied in the upper atmosphere for many years, and its
presence has been measured. Free oxygen atoms are formed in the upper atmosphere
John H. Seinfeld and Spyros N. Pandis, Atmospheric Chemistry and Physics : From Air Pollution to
Climate Change, A Wiley-Interscience Publication (New York: Wiley, 1998), chap. 4,
http://lynx.lib.usm.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=nlebk&AN=261
30&site=ehost-live.
15
U. Kogelschatz, B. Eliasson, and M. Hirth, “Ozone Oeneration from Oxygen and Air: Discharge Physics
and Reaction Mechanisms,” 1988.
14
8
when solar ultra-violet radiation (UV) with wavelengths less than 242 nm dissociates
molecular oxygen. This equation is:
𝑂2 + ℎ𝜈 → 𝑂 + 𝑂
(3.2)
where ℎ𝜈 is the energy of the UV photon. Once free oxygen atoms are available, ozone
can be created. Once created, ozone is a very strong optical absorber of wavelengths 240
nm to 320 nm. The photon’s energy can dissociate ozone to the following:
𝑂3 + ℎ𝜈 → 𝑂2 + 𝑂.
(3.3)
Equation (3.3) is the reason for the scarcity of this wavelength band in the lower
atmosphere, and is an indication of the significant amounts of ozone in the upper
atmosphere 16. Ozone can also decompose when reacted with another free oxygen atom
to form two oxygen molecules:
𝑂3 + 𝑂 → 𝑂2 + 𝑂2 .
(3.4)
Equations (3.2), (3.3), and (3.4) were proposed in 1930 by Sydney Chapman to explain
the production and destruction of ozone in the upper atmosphere, and is still known as the
Chapman mechanism. This mechanism was the dominant conceptual model until the
mid-1960s when measurements of the upper atmosphere showed that ozone levels were a
factor of two smaller than predicted. This demonstrated that significant ozone
destruction reactions were occurring other than the single ozone destruction process in
the Chapman mechanism. In 1970, Paul Crutzen pioneered the discovery of the role of
nitrogen oxides in stratospheric chemistry. The ozone destruction processes that had to
be added to the Chapman mechanism have the form of the following catalytic cycle:
16
Julian Heicklen, Atmospheric Chemistry (Academic Press, 1976); Seinfeld and Pandis, Atmospheric
Chemistry and Physics, chap. 4.
9
𝑋 + 𝑂3 → 𝑋𝑂 + 𝑂2
(3.5𝑎)
𝑋𝑂 + 𝑂 → 𝑋 + 𝑂2
(3.5𝑏)
Net: 𝑂3 + 𝑂 → 𝑂2 + 𝑂2
(3.5𝑐)
where 𝑋 is a free radical catalyst and can be either nitrogen monoxide (𝑁𝑂), hydrogen
(𝐻), hydroxide (𝑂𝐻), chlorine (𝐶𝑙), or bromine (𝐵𝑟). The catalyst, by definition, is not
consumed in the process, and the net effect results in the reaction of ozone and atomic
oxygen forming two oxygen molecules 17. Because this research was performed within a
controlled vacuum chamber of only nitrogen and oxygen mixtures, the NO catalytic cycle
is the only catalytic cycle that will be discussed. Below is the catalytic cycle of NO in
the upper atmosphere:
𝑁𝑂 + 𝑂3 → 𝑁𝑂2 + 𝑂2
(3.6𝑎)
𝑁𝑂2 + 𝑂 → 𝑁𝑂 + 𝑂2
(3.6𝑏)
Net: 𝑂3 + 𝑂 → 𝑂2 + 𝑂2
(3.6𝑐)
where 𝑁𝑂2 is nitrogen dioxide. The net result is the conversion of two odd oxygen
species (O3, O) to two even oxygen species (O2, O2). In the lower stratosphere where
ozone is more prevalent, and when sunlight is available, another NO catalytic cycle
exists:
17
𝑁𝑂 + 𝑂3 → 𝑁𝑂2 + 𝑂2
(3.7𝑎)
𝑁𝑂2 + 𝑂3 → 𝑁𝑂3 + 𝑂2
(3.7𝑏)
𝑁𝑂3 + ℎ𝜈 → 𝑁𝑂 + 𝑂2 or 𝑁𝑂2 + 𝑂
(3.7𝑐)
Net: 2 𝑂3 → 3 𝑂2
(3.7𝑑)
Seinfeld and Pandis, Atmospheric Chemistry and Physics, chap. 4.
10
where 𝑁𝑂3 is nitrogen trioxide. Equation (3.7𝑐) has two possible reactions with the first
product being eight times less likely, and the first possibility results in the conversion of
odd oxygen to even oxygen. One more NO catalytic cycle results in the destruction of
ozone into an even oxygen molecule and an odd oxygen atom:
𝑁𝑂 + 𝑂3 → 𝑁𝑂2 + 𝑂2
(3.8𝑎)
𝑁𝑂2 + ℎ𝜈 → 𝑁𝑂 + 𝑂
(3.8𝑏)
Net: 𝑂3 + ℎ𝜈 → 𝑂2 + 𝑂.
(3.8𝑐)
The above cycle is referred to as a null cycle because odd oxygen is not destroyed, but
the reactions of this cycle were relevant for this research 18. The first and second
equation of this cycle help explain the results of nitrogen oxides and ozone studies that
correlated to the morning and afternoon rush hours in the Meadowlands of New Jersey.
During the morning rush hour, an increase of NO from vehicle emissions was evident.
The morning’s rapid increase of NO resulted in a dip of background ozone levels because
new NO scavenged the background O3 to make NO2. When light of wavelengths shorter
than 400 nm became available from the morning sun, the dissociation of NO2 to NO and
O resulted in O3. O3 continued to grow even after the morning rush hour stopped
producing NO because enough NOx was now present for O3 formation. O3 eventually
leveled out once the reservoir of NOx had been utilized in O3 production. When the
evening rush hour started, the new production of NO again scavenged the background O3
to make NO2 resulting in a drop of O3. As the sun went down, O3 levels continued to
decrease while NO2 levels increased because of the absence of the dissociation reaction
18
Ibid.
11
of NO2 with sunlight. These facts reinforced the need for NO2 to photodissociate in order
to generate ozone in the lower atmosphere 19. The dissociation of NO2 is also a possible
explanation for effects observed in the measurements of ozone during the research
discussed here.
One more reaction,
𝑁𝑂3 + 𝑁𝑂2 + 𝑀 ↔ 𝑁2 𝑂5 + 𝑀
(3.9)
where 𝑁2 𝑂5 is dinitrogen pentoxide, will be mentioned because of its possible relevance
to this experiment 20. NO2 and NO3 have a role in the creation and destruction of ozone.
Ozone Formation from Ionizing Radiation
A number of studies have been performed on the chemical effects of radiation on
air-like mixtures. Different forms of ionizing radiation have been studied which include
alpha, beta, and gamma interaction with N2-O2 mixtures.
In 1929, alpha interactions were studied by flowing N2-O2 mixtures over an alpha
ray bulb system and measuring the chemicals produced using chemiluminescence.
Levels of ozone, nitrous acids, and nitrogen monoxide were measured. The presence of
nitrous acid suggests the presence of hydrogen in the experiment 21.
Beta radiation has been simulated in N2-O2 mixtures with the use of high energy
electron beams. In 1970, a UV absorption method was used to measure ozone in N2-O2
mixtures as a result of the ionizing radiation from the electron pulses of a Febetron 705.
Once nitrogen and oxygen ions were formed from the ionizing radiation, reactions could
Dawn Roberts–Semple, Fei Song, and Yuan Gao, “Seasonal Characteristics of Ambient Nitrogen Oxides
and Ground–level Ozone in Metropolitan Northeastern New Jersey,” Atmospheric Pollution Research 3,
no. 2 (April 2012): 247–57, doi:10.5094/APR.2012.027.
20
Seinfeld and Pandis, Atmospheric Chemistry and Physics, chap. 4.
21
Lind and Bardwell, “Ozonization and Interaction of Oxygen with Nitrogen under Alpha Radiation.”
19
12
occur that resulted in atomic oxygen. These nitrogen and oxygen ions could also form
different nitrous oxides. Because atomic oxygen production is required for ozone
production, without the presence of any nitrous oxide, free oxygen was believed to be
formed from two ion reactions and one neutral reaction:
𝑂2+ + 𝑒 → 2𝑂
(3.10𝑎)
𝑂2+ + 𝑂2− → 2𝑂 + 𝑂2
(3.10𝑏)
𝑁 + 𝑂2 → 𝑁𝑂 + 𝑂
(3.10𝑐)
where 𝑂2+ and 𝑂2− are oxygen ions and 𝑒 is an electron. Because nitrous oxides are
formed quickly due to the exposure of ionizing radiation, free oxygen atoms can be
formed by several reactions involving nitrous oxide ions and neutrals:
𝑁𝑂+ + 𝑒 → 𝑁 + 𝑂
(3.11𝑎)
𝑁 + 𝑁𝑂 → 𝑁2 + 𝑂
(3.11𝑏)
𝑁 + 𝑁𝑂2 → 𝑁2 𝑂 + 𝑂
(3.11𝑐)
𝑁 + 𝑁𝑂2 → 𝑁2 + 2𝑂
(3.11𝑑)
where 𝑁𝑂+ is the nitrogen oxide ion. The computational models for ozone production
performed using their available rate constants did not completely agree with their
measurements 22. In 1984, measurements of 80%/ 20% nitrogen/ oxygen mixtures were
performed to show the dependence of ozone production on direct electron dissociation of
oxygen. Results showed that roughly half of the ozone produced resulted from direct
electron dissociation of oxygen, while the other half of ozone was formed from nitrogen
C. Willis, A. W. Boyd, and M. J. Young, “Radiolysis of Air and Nitrogen-Oxygen Mixtures with Intense
Electron Pulses: Determination of a Mechanism by Comparison of Measured and Computed Yields,”
Canadian Journal of Chemistry 48, no. 10 (1970): 1515–1525.
22
13
atoms or excited nitrogen molecules. The 1984 paper did add reactions that could create
free oxygen atoms that were not included in the 1970 paper. The new reactions are:
𝑁2∗ + 𝑂2 → 𝑁2 + 𝑂2∗ → 𝑁2 + 2𝑂 or 𝑁2 𝑂 + 𝑂
(3.12)
where 𝑁2∗ and 𝑂2∗ are excited molecules 23.
The production of ozone in the presence of gamma radiation was studied in 2014
using the University of New Mexico’s AGN-201 M reactor. Ozone levels that were
smaller than the levels of atmospheric background ozone were measured as a result of
radiation interactions 24.
Studies have also been performed showing that the inclusion of water in irradiated
nitrogen-oxygen mixtures leads to free hydrogen atoms. These free hydrogen atoms can
produce a number of acids, and these acids have been measured 25.
This research differs from the above studies because of the improvement of
technology from that time. Measurements can be made on smaller time scales in order to
better understand the chemical kinetics. The use of the cycled radiation cover also allows
further evidence to help validate a model currently under development by this research
group. The inclusion of the dye laser will allow future students to measure different
products that will contribute to validating the model.
B. Eliasson, U. Kogelschatz, and P. Baessler, “Dissociation of O2 in N2/O2 Mixtures,” Journal of
Physics B: Atomic and Molecular Physics 17, no. 22 (1984): L797.
24
J. Cole et al., “Radiolytic Yield of Ozone in Air for Low Dose Neutron and X-Ray/gamma-Ray
Radiation,” Radiation Physics and Chemistry 106 (January 2015): 95–98,
doi:10.1016/j.radphyschem.2014.06.008.
25
A. Russell Jones, “Radiation-Induced Reactions in the N2-O2-H2O System,” Radiation Research 10, no.
6 (June 1, 1959): 655–63, doi:10.2307/3570649.
23
14
CHAPTER IV – CAVITY ENHANCED ABSORPTION SPECTROSCOPY
Absorption spectroscopy is based on the fact that each atom and molecule
interacts differently with electromagnetic radiation. This principle can be used to identify
the presence of a specific atom or molecule (species) and can also identify the
concentration of a specific species when that species’ absorption spectrum is known. An
absorption spectrum is a measure of the atom or molecule’s ability to absorb photons at a
variety of wavelengths. Each atom and molecule has a unique absorption spectrum
because of the different number of elementary particles and physical configurations for
the atomic or molecular electrons. This results in the species having a variety of excited
states, while molecules are also capable of mechanical excitations (twisting, bending,
vibrating, rotating, etc.). Because of this, atoms and molecules absorb photons with
energies equal to their excitation energies 26. Figure 1 shows the absorption spectrum for
ozone. The names on the figure reflect the discoverer of the different absorption bands.
W. G. Richards and P. R. Scott, Structure and Spectra of Molecules, 1 edition (Chichester West Sussex ;
New York: Wiley, 1985), chap. 1.
26
15
Figure 1. Ozone Absorption Spectrum.
This figure has been reproduced with permission (See Appendix A) 27.
A species’ ability to absorb a particular wavelength has been experimentally
quantified using absorption cross sections over different wavelength regions. The
absorption cross sections are typically given in cm2/molecule and act as an effective area
to the photons for interaction and absorption. The larger valued cross sections represent
wavelengths of light that will be absorbed more than the smaller valued cross sections.
The amount of light that will be absorbed can be calculated using Beer’s Law, which
states that the intensity of a beam of light is diminished by a factor which depends on the
Hannelore Keller-Rudek et al., “The MPI-Mainz UV/VIS Spectral Atlas of Gaseous Molecules of
Atmospheric Interest,” n.d., http://www.uv-vis-spectral-atlas-mainz.org/; M Ackerman, “UV-Solar
Radiation Related to Mesospheric Processes,” in Mesospheric Models and Related Experiments, ed. G
Fiocco (Dordrecht: D. Reidel Publishing Company, 1971), 149–59.
27
16
number density and wavelength-dependent cross section of the interacting atoms or
molecules in a given path length. Below is Beer’s Law:
𝐼(𝑥, 𝜆) = 𝐼0 (𝑑)𝑒 −𝛼(𝜆)𝑥
(4.1)
given 𝛼(𝜆) = 𝜎(𝜆)𝑁
where 𝐼 is the detected intensity of light of wavelength 𝜆 after traveling a distance 𝑥 in
meters through the presence of the absorber, 𝐼0 is the initial intensity of light, 𝑑 is total
distance of light traveled before reaching the sample, 𝛼(𝜆) is the attenuation coefficient,
𝜎(𝜆) is the absorption cross section in meter2 per molecule, and 𝑁 is the number density
of the absorber in number per meter3 28.
Figure 2. Basic representation of Beer’s Law.
The blue cylinder represents an absorbing medium.
One can use Beer’s Law to construct an absorption spectrum by exposing an
unknown sample to a continuous distribution of light. By scanning the transmittance at
wavelength steps and comparing incident intensity to transmitted intensity, an absorption
spectrum can be constructed. Beer’s Law can also be used to calculate the number
density of an absorber if the cross section is known for that absorber. The only problem
with Beer’s Law is when low number densities or weak absorbers are being measured,
the difference in incident and transmitted intensity is too low to be measured. The only
Barbara A Paldus and Alexander A Kachanov, “An Historical Overview of Cavity-Enhanced Methods,”
Canadian Journal of Physics 83, no. 10 (October 2005): 975–99, doi:10.1139/p05-054; Reese, “A System
for Measuring Radiation Induced Chemical Products in Atmospheric Gases Using Optical Detection
Methods,” chap. 4.
28
17
way to change the fractional change in intensities is to increase the path length through
the absorber. Cavity enhanced absorption spectroscopy (CEAS) is used to greatly
increase the effective path length within the laboratory environment.
Figure 3. Basic representation of CEAS.
A basic CEAS system uses two highly reflective mirrors to redirect the incoming
light that makes it into the cavity several times through the absorbing sample. This
increased effective path length allows an extended absorption to occur. The optical
cavity used for this type of research utilizes two spherical mirrors separated by a distance
d. The mirrors are assumed to have equal reflectivity, R, and transmit a small percentage,
T. Both reflectivity and transmittance are functions of wavelength. With the cavity
mirrors aligned so no light can diverge out of the cavity before being absorbed or
transmitted through the mirrors, an effectively long path length can be created. Each time
the light reflects off of a cavity mirror, the transmitted light is reduced by a factor of 1-R.
For the light to reach a detector on the other side of the cavity, the incoming light would
reduce by a factor of T at the mirror nearest the light source. When the light reaches the
second cavity mirror, the intensity would reduce by another factor of T before reaching
the detector with an overall reduction in intensity of T2 after traveling a distance d.
Beer’s Law after making one pass through two mirrors would be:
18
𝐼 = 𝐼0 𝑇 2 𝑒 −𝛼(𝜆)𝑑 .
(4.2)
Light in the cavity for one round-trip would experience an overall intensity reduction of
R2T2 after traveling a cavity distance of 2d. Beer’s Law after making one round-trip
would be:
𝐼 = 𝐼0 𝑅 2 𝑇 2 𝑒 −𝛼(𝜆)2𝑑 .
(4.3)
Light that makes two round-trips would experience an overall intensity reduction of R4T2,
or R2(2)T2, after traveling a cavity distance of 4d, or 2(2)d. Beer’s Law after making two
round-trips would be:
𝐼 = 𝐼0 𝑅 4 𝑇 2 𝑒 −𝛼(𝜆)4𝑑 .
(4.4)
After making n round-trips within the cavity before hitting the detector, the intensity
would be reduced by a factor of R2nT2 after traveling a cavity distance of (2n)d. Beer’s
Law after making 𝑛 round-trips would be:
𝐼 = 𝐼0 𝑅 2𝑛 𝑇 2 𝑒 −𝛼(𝜆)2𝑛𝑑 .
By rearranging, 𝐼 = 𝐼0 𝑇 2 𝑒 2𝑛 𝑙𝑛(𝑅) 𝑒 −𝛼(𝜆)2𝑛𝑑 .
(4.5)
When an absorber is not present, α(λ) = 0, the second exponential term vanishes. With
no absorber, all intensity lost within the cavity is due solely to losses from reflections.
By rearranging equation (4.5) with no absorber, the intensity drops to a value of
𝐼 = 𝐼0 𝑇 2 (1⁄𝑒)
(4.6𝑎)
when 𝑛 = − 1⁄2𝑙𝑛𝑅 .
(4.6𝑏)
This means that without an absorbing sample, the number of round-trips required to
reduce the intensity by 1/e would depend only on the reflectivity of the mirrors.
19
Many different methods utilize CEAS. The method used in this research is cavity
ringdown spectroscopy (CRDS) using a pulsed laser as the light source. For this method,
instead of measuring initial and final intensities, the decay constant of the intensity of the
cavity output is measured. The rate of the decay depends on mirror reflectivity, mirror
separation distance, and the concentration of the absorbing sample.
An equation for this exponential decay can be reached from equation (4.5) along
with the time it takes for light to travel n round-trips in the cavity. The time, 𝑡, it takes
for the light to travel is equal to the distance traveled after n round-trips, or 2𝑛𝑑, divided
by the speed of light, 𝑐:
𝑡 = 2𝑛𝑑⁄𝑐 ,
or
𝑡 (𝑐⁄2𝑛𝑑 ) = 1.
(4.7)
By rearranging equation (4.7) and using equation (4.5):
𝐼 = 𝐼0 𝑒 2𝑛(𝑙𝑛𝑅−𝛼(𝜆)𝑑)(
𝐼 = 𝐼0 𝑒
𝑡𝑐⁄
2𝑛𝑑)
−𝑡𝑐(−𝑙𝑛𝑅+𝛼(𝜆)𝑑)⁄
𝑑
or
𝑡
𝐼 = 𝐼0 𝑒 − ⁄𝜏
(4.8𝑎)
𝑐[−𝑙𝑛𝑅 + 𝛼(𝜆)𝑑]⁄
where 1⁄𝜏 =
𝑑.
(4.8𝑏)
The 𝑇 2 factor is assumed to be a part of 𝐼0 since CRDS involves only the time
dependence of the cavity output. 𝜏 is known as the cavity ringdown time, which is the
20
time it takes for the output intensity to reduce by a factor of 1/e 29. A characteristic
ringdown time, τ0, is measured from the time it takes for the loss through the mirrors with
no absorbers, α(λ) = 0, to reduce the output intensity by 1/e:
1⁄ = 𝑐(−𝑙𝑛𝑅)⁄ .
𝜏0
𝑑
(4.9)
The above equation can be solved for the effective reflectivity of the cavity mirrors:
𝑅 = exp(− 𝑑⁄𝑐𝜏0 ) .
(4.10)
Once an absorber with a known absorption spectrum is introduced, the ringdown time
represents a combination of light being absorbed and light transmitting through the
mirrors. To calculate the number density of the absorber with a known absorption
spectrum, use equation (4.8𝑏) and equation (4.9):
1⁄ − 1⁄ = 𝑐𝛼(𝜆)
𝜏
𝜏0
(4.11𝑎)
where 𝛼(𝜆) = 𝜎0 (𝜆)𝑁0 + 𝜎1 (𝜆)𝑁1 + … + 𝜎𝑠 (𝜆)𝑁𝑠
(4.11𝑏)
and 𝑠 is the number of absorbers with a cross section at wavelength 𝜆. When one
absorber is being measured, equation (4.11𝑎) can be rewritten as:
1⁄ − 1⁄ = 𝑐𝜎(𝜆)𝑁
𝜏
𝜏0
𝑁 = 1⁄𝑐𝜎(𝜆) (1⁄𝜏 − 1⁄𝜏0 ).
(4.12)
To solve for the concentration of an absorber, the number density can be converted to
parts per million (ppm) or parts per billion (ppb) by dividing the number density of the
absorber by the number density solved from the ideal gas law 30,
George Miller and Christopher Winstead, “Cavity Ringdown Laser Absorption Spectroscopy,” in
Encyclopedia of Analytical Chemistry, ed. R.A. Meyers (Chichester, England: John Wiley and Sons, Ltd.,
2000), 10734–50.
30
Reese, “A System for Measuring Radiation Induced Chemical Products in Atmospheric Gases Using
Optical Detection Methods,” chap. 4.
29
21
𝑁 = 𝑃⁄𝑘𝑇 ,
(4.13)
and then multiplying the result by 1.0 x 106 or 1.0 x 109 for ppm or ppb. Above, 𝑃 and 𝑇
represent the pressure and temperature of the chamber in Pascal and Kelvin, and 𝑘 is the
Boltzmann constant.
In a typical configuration, CRDS is orders of magnitude more sensitive than
single pass absorption measurements. This level of sensitivity is important for
experiments where concentrations of important species are anticipated to be at the partsper-billion level. CRDS also has the ability to make absolute concentration
measurements without the need for complex calibration schemes. If the absorption crosssection for a given species is known at a specific wavelength, and the reflectivity (or
effective reflectivity that combines all other losses) of the mirrors can be determined
without that species present, the number density of that species can be determined using
equation (4.12). For example, with the radiation source shielded, an effective reflectivity
can be determined that includes the reflectivity losses of the mirrors, as well as losses due
to absorption and scattering from any background species in the gas. When the radiation
source is exposed, the change in ringdown time can be used to measure the concentration
of a species induced by the presence of the radiation.
22
CHAPTER V – EXISTING SYSTEMS TO BE UTILIZED
Prior to this project, there was a project in the laboratory to create a system for
measuring radiation-induced chemical products using Cavity Ringdown Spectroscopy
(CRDS) 31. That research explored the potential for using CRDS to evaluate the presence
of chemical products generated by air-radiation interactions in an existing vacuum
chamber. A brief summary of this existing system is provided prior to discussion of
system modifications.
Existing Vacuum System
The existing vacuum system was originally created as a means to explore
radiation-induced fluorescence. With future work in mind, the vacuum system was
originally designed with the capability of simulating various atmospheric conditions 32.
The vacuum system was later modified to explore Cavity Ringdown Spectroscopy with
the addition of the inner chamber cavity 33. Figure 4 is a diagram that shows the location
and use of each port before providing further detail.
Reese, “A System for Measuring Radiation Induced Chemical Products in Atmospheric Gases Using
Optical Detection Methods.”
32
Aubri Capri Buchanan, “Characterization of Air Fluorescence Induced by Alpha Radiation” (Master’s
Thesis, The University of Southern Mississippi, 2008).
33
Reese, “A System for Measuring Radiation Induced Chemical Products in Atmospheric Gases Using
Optical Detection Methods.”
31
23
Figure 4. Vacuum chamber port use and location diagram.
Physical Characteristics
The vacuum system utilizes an 18” diameter stainless steel spherical chamber
with 19 circular access ports of varying flange sizes ranging from 2.75” to 8”. Four of
24
the 2.75” ports are used for attaching temperature and pressure sensors. One 2.75” port
uses a vent valve to return the chamber to ambient pressure. One 2.75” port has a valve
opening to a controlled manifold that can supply precise gas mixtures. The manifold
consists of three gas introduction valves, a metering valve, a bypass valve, and a shutoff
valve. One 2.75” port is connected to an angle valve to expose the chamber to a small
roughing pump. The bottom 8” port is connected to a gate valve to expose the chamber
to the diffusion pump and larger roughing pump. An 8” port on the side has a sealablehinged door for access into the chamber. Four opposing ports can be used as optical
axes. One of these optical paths is used and described in the upcoming CRDS system.
Figure 5. Previously developed vacuum system.
Figure 5 was taken after the vacuum system was moved to a different lab than the one in which it was originally developed. Figure 5
was taken during the process of rotating the chamber 135 degrees as described in Chapter VI.
25
Two of the six remaining ports are used to mount and shield/expose a radiation
source. The top 8” port uses stainless steel all thread to suspend the aluminum radiation
source holder. The aluminum radiation source holder can hold one or two 1.25” diameter
radiation sources. One 2.75” port is used to mount the pneumatic linear feedthrough.
The pneumatic linear feedthrough with 2” of travel is used to slide an aluminum plate to
cover and uncover the radiation sources without opening the chamber. The linear
feedthrough is actuated by a 4-way solenoid valve.
Figure 6. Aluminum source holder and radiation source cover.
Source cover, radiation sources, and thermocouples were removed for the picture on the left.
26
Figure 7. Polonium 210 Radiation Source
Five mCi
Pumping Mechanisms
Three pumps are used in the vacuum system: a diffusion pump, and two rotary
vane pumps. Both mechanical rotary vane pumps are connected to the vacuum system
with angle valves to isolate the system from the pumps when not in use. The smaller
roughing pump is connected directly to the chamber and is used to evacuate the chamber
from ambient pressure to below 200 mTorr. The larger roughing pump is connected to
the foreline of the diffusion pump. The diffusion pump is connected to the 8” port on the
bottom of the chamber with a gate valve. Once the diffusion pump is warmed up and the
chamber is below 200 mTorr, the smaller roughing pump is closed and the gate valve is
opened. With the diffusion pump, the vacuum system can reach high vacuum pressures
in the 1.0 x 10-7 to 1.0 x 10-6 Torr range.
Chamber Measurement Capabilities
The chamber’s pressure is monitored using three gauges: an ion gauge, and two
capacitance manometers. The upper capacitance manometer measures from 1000 Torr to
1 Torr and the lower capacitance monometer measures from 1 Torr to .001 Torr. The ion
gauge can measure from .005 Torr to 1.0 x 10-10 Torr. Two thermocouples on a 2.75”
thermocouple flange are used to monitor the inner chamber wall and the chamber
27
atmosphere temperature. All gauges and thermocouples are monitored using an NI
CompactRIO (cRIO), where RIO stands for reconfigurable I/O modules. Pressure and
temperatures are displayed on the vacuum cart’s push button control panel and can be
sent to a computer for recording.
System Control
All of the vacuum system’s valves and gauges are monitored and operated by a
cRIO. The cRIO is an embedded controller with a real time processor for communication
and signal processing. The cRIO uses I/O modules to connect to all of the vacuum
system’s components.
cRIO
Figure 8. Old cRIO.
The pictured cRIO was replaced after a power cycle damaged the internal memory. The cRIO was replaced by a newer model and the
damaged cRIO was fixed and is available for use.
The cRIO also controls a push button panel that allows the user to control the
system and view its status without needing a computer. A computer can be connected to
access the controller via a graphical user interface (GUI) 34.
34
Ibid.; Buchanan, “Characterization of Air Fluorescence Induced by Alpha Radiation.”
28
Existing CRDS System
The optical system originally developed for CRDS utilizes two 2” optical mounts
and two custom machined 8” flanges to mount the CRD cavity in the vacuum chamber.
The optical mounts are mounted to the interior side of the 8” flanges using brackets.
Mounted at the center of the 8” flanges are 1.33” quartz viewport flanges, and mounted
on either side of the viewports are two rotary feedthroughs aligned with the optical mount
adjusters. The ends of the rotary feedthroughs are coupled into the slotted brass fittings
of the x and y adjustment of the optical mounts. This allows adjustments of the optical
cavity while the chamber remains sealed.
Figure 9. Previously modified 8” flange for CRD cavity.
A large angle iron bracket that was installed on the exterior of the chamber is used
to hold optical elements mounted to a breadboard. The bracket was installed directly on
29
the chamber to eliminate any vibrations that might be experienced independent of the
vibrations the entire chamber experiences. A small 20 Hz pulsed Neodymiun-doped
yttrium aluminum garnet (ND: YAG) laser outputting 266 nm UV was used along with
apertures to turn and shape the input beam in the original experiments.
Figure 10. Previously used laser configuration.
This picture was reproduced with permission (See Appendix A) 35.
LabVIEW alignment software and waveform analysis software were developed
and have since been merged and modified. Using cavity mirrors that were 99.7%
reflective at 266 nm, one standard deviation from average was equal to 1.6% of τ0. One
Reese, “A System for Measuring Radiation Induced Chemical Products in Atmospheric Gases Using
Optical Detection Methods,” 35.
35
30
standard deviation corresponded to an error in ozone concentration of 6.01 ppb. A
photomultiplier tube (PMT) and National Instruments (NI) PXI data acquisition system
was, and still is used to monitor the PMT signal of every laser pulse 36.
Figure 11. PXI used for data acquisition.
36
Ibid., chap. 6.
31
CHAPTER VI – SYSTEM MODIFICATIONS
To incorporate the current vacuum system and CRDS system with a tunable dye
laser system, a stand-alone optics table was required to support the weight of the laser
system. Modifications of the vacuum system were required to align the dye laser’s
output with the input of the CRDS cavity. To measure the initial concentrations of the
gas sample in the vacuum chamber, a quadrupole mass spectrometer was implemented.
Laser System
For access to wavelengths from 200 nm –700 nm, a Sirah dual-grating (2400
g/mm gratings) high-resolution pulsed dye laser can be used. The dye laser is pumped by
a Spectra Physics Nd: YAG laser. The dye laser system was purchased on an earlier
project.
Figure 12. Dye Laser and Nd: YAG laser.
To support the weight and size of the dye laser and Nd: YAG laser, a 4’ x 8’
Newport optics table (RS 4000) was used. The optics table was aligned parallel with the
vacuum system’s cart. The vacuum chamber had to be rotated 135 degrees to align the
CRDS cavity axis perpendicular to the dye laser’s output. A second external optics
bracket was installed for use of optics on both sides of the vacuum chamber. Using
32
irises, a turning mirror, and a vertical shift apparatus that uses two more turning mirrors,
the dye laser can be guided into the CRDS cavity. Different turning mirrors were used as
filters because of their reduced reflectivity to reduce the laser’s input intensity to the
cavity. This was an alternative to using many neutral density filters on the PMT and
helped to reduce the intensity in the cavity to avoid saturation effects. The small Nd:
YAG can be placed between the dye laser’s first turning mirror and the vertical shift
apparatus for use of 266 nm, 532 nm, and 1064 nm harmonic wavelengths.
To Chamber
Figure 13. Small ND: YAG in use.
Table 2 shows the turning mirrors used at each different wavelength during this
project.
33
Table 2
Mirror Configurations
Upper Telescope
Mirror
Lower Telescope
Mirror
266 nm Rayleigh
Scattering
1” Y3 (532 nm)
1” Y3
266 nm SO2
1” Y3
1” Y3
283 nm SO2
1” Y4 (266 nm)
1” Y4
266 nm O3
1” Y3
1” Y3
Turning Mirror
Following Dye
Laser
1” Y3
Two 1” 266 nm CRD mirrors with 1 m radius of curvature were used for all experiments.
Universal Gas Analyzer (UGA)
The Stanford Research Systems (SRS) UGA200 was installed to analyze the
initial concentrations of the vacuum chamber’s sample gas. A table was made to elevate
the UGA above the vacuum chamber at an adjustable height. Rubber was placed
between the tabletop and the frame to reduce any small vibrations from the UGA.
34
Figure 14. UGA table with adjustable height.
A stainless steel capillary was used for the inlet of the UGA. The capillary can be
used to probe above the center of the chamber at a variable height. The capillary was fed
into the chamber using a small vacuum feedthrough mounted on a 2.75” flange which
was mounted on an 8” flange. The capillary provides a pressure drop to introduce the
sample into an intermediate vacuum chamber in the UGA. The gas from this chamber is
then directed to a quadrupole mass spectrometer at ultra-high vacuum to provide a mass
measurement. The exhaust of the UGA was sent to the vent hood along with the exhaust
from both roughing pumps.
35
Figure 15. Swagelok stainless steel Ultra-Torr fitting used for UGA capillary.
Swagelok Part Number: SS-1-UT-1-2
An example spectra was recorded using the factory’s provided software and is
shown in Figure 16. A background subtraction was performed, which included scanning
the UGA’s vacuum chamber with the quadrupole mass spectrometer and then subtracting
the UGA’s vacuum chamber spectrum from the recorded spectra of interest. A 12 scan
average from 0 amu – 65 amu is shown below with the Analyze Function to analyze
cracking patterns of a given gas library based on the effect of the UGA’s filament.
Carbon Monoxide was unselected in the Analyze Function’s library.
36
Figure 16. UGA mass spectrum
The Analyze Function results are shown in a separate window in the upper right corner of the figure.
Measurement Device
A photomultiplier tube was used to measure the cavity output intensity, but
different optical filters were used prior to the PMT to select different wavelengths needed
for this project.
37
Figure 17. PMT used in 266nm configuration.
Neutral density filters had to be used to reduce the cavity output intensity. Using
neutral density filters to reduce signal intensity is not the best method to prevent PMT
saturation. If the PMT has random noise variations that are generating apparent
fluctuations in the signal, filtering out half the light would make the same noise
fluctuations twice as big relative to the signal. An alternative method to reducing
saturation is to reduce the driving voltage to the PMT. However, below a minimum
voltage the response of the PMT can be non-linear, so neutral density filters are required
additionally. As mentioned above, different turning mirrors or beam splitters were used
to reduce the input intensity before this point. A piece of frosted quartz was used right
before the PMT face to diffuse the light to avoid high intensity spots that could
overexcite small regions of the PMT. Table 3 shows the filters used for each different
experiment during this project.
38
Table 3
PMT Filter Configurations
Adapter 1
Adapter 2
Adapter 3
Adapter 4
2” O.D. 0.5
ND Filter
2” O.D. 0.5
ND Filter
266 nm
Rayleigh
Scattering
1” Frosted
Glass (with 2”
to 1” adapter)
2” 280 nm +/12.5 nm
Bandpass Filter
2” Optical
Density (O.D.)
1.0 Neutral
Density (ND)
Filter
266 nm SO2
1” Frosted
Glass
2” 280 nm +/12.5 nm
Bandpass Filter
2” O.D. 1.0
ND Filter
283 nm SO2
1” Frosted
Glass
2” 280 nm +/12.5 nm
Bandpass Filter
2” O.D. 1.0
ND Filter
266 nm O3
1” Frosted
Glass
2” 280 nm +/12.5 nm
Bandpass Filter
2” O.D. 1.0
ND Filter
39
2” O.D. 0.5
ND Filter
CHAPTER VII – SOFTWARE DEVELOPMENT
The cRIO, dye laser, mass spectrometer, and PXI chassis are controlled using
LabVIEW. LabVIEW, short for Laboratory Virtual Instrument Engineering Workbench,
is a graphical programming language created by NI. A LabVIEW program is made of
one or more virtual instruments (VI). The term virtual instrument is used because the
program’s appearance and operation often imitate actual physical instruments. However,
VIs are analogous to functions, subroutines, and main programs from other popular
programming languages, like C. VIs have a front panel and a block diagram. The front
panel is the interactive user interface of the VI, and simulates the front panel of a physical
instrument. The block diagram is the VI’s source code 37.
The NI cRIO-9066 and NI PXI 1036 chassis are programmed using LabVIEW.
The Sirah dye laser and SRS UGA200 mass spectrometer were purchased with LabVIEW
drivers. The provided LabVIEW drivers were used and modified to create separate VIs
that could independently control the instruments for our needs without needing the
factory’s provided software package. Most importantly, variables could be shared among
our VIs for data collection and automation purposes. Variables shared among computers
use LabVIEW’s network-published shared variables. “Network-published shared
variables publish data over a network through LabVIEW’s Shared Variable Engine
(SVE). The SVE manages shared variable updates using the NI Publish-Subscribe
Protocol (NI-PSP). The term publish-subscribe describes a model of communication
where writers, or publishers, do not send data to specific readers, or subscribers. Instead,
37
Jeffrey Travis and Jim Kring, LabVIEW for Everyone: Graphical Programming Made Easy and Fun, 3rd
ed (Upper Saddle River, NJ: Prentice Hall, 2007).
40
publishers send updates to a server, in this case the SVE, and subscribers receive those
updates from the server 38.”
Laboratory Network
The experimental setup includes two offline Windows XP computers and an
online Windows 7 computer networked together using a wireless router. Both Windows
XP desktops are hard-wired to the router and the Windows 7 desktop connects wirelessly.
The cRIO, dye laser, and mass spectrometer connect to a Windows XP computer
running LabVIEW 2014 SP1. The NI cRIO-9066 connects using a crossover ethernet
cable. The dye laser connects using a straight-through female-to-female RS232 serial
cable and a camera cable requiring an EPIX D2X1 PCI card. The EPIX D2X1 PCI card
is not compatible with Windows 7. The SRS UGA200 mass spectrometer connects using
a straight-through male-to-female RS232 serial cable. The factory provided SRS
UGA200 software is not compatible with Windows XP. The LabVIEW VI “UGA.vi”
must be used to operate the UGA from the Windows XP computer. A USB to Ethernet
adapter and serial PCI express card are required for an extra Ethernet port, and two serial
ports.
The NI PXI 1036 chassis connects to the second Windows XP computer running
LabVIEW 2009 SP1 using a NI PXI-8310 card. The NI PXI-8310 card is not compatible
with Windows 7 and uses two ethernet cables and a notebook card that requires a PCI to
PCMCIA adapter card. LabVIEW 2009 SP1 is used instead of LabVIEW 2014 SP1
because LabVIEW 2014 SP1 cannot record data at a real time rate when run together
“Understanding Shared Variable Technology - LabVIEW 2011 Help - National Instruments,” accessed
September 16, 2016, https://zone.ni.com/reference/en-XX/help/371361H-01/lvconcepts/ni_psp/.
38
41
with Windows XP, the NI PXI-8310 card, and the NI PXI 1036 chassis. The NI PXI5124 digitizing module is used in the PXI chassis to monitor the PMT signal of every
laser pulse. The Spectra Physics ND: YAG laser is also connected to the same computer
using a straight-through male-to-female RS232 serial cable. The laser’s GCR software is
available, but rarely used, for reading interlock faults and helping diagnose laser failures.
The final computer, running Windows 7 with LabVIEW 2014 SP1, is not
necessary. The Windows 7 computer is used to remote into the other two computers
from the lab office for total control from one computer and to send notification emails to
the user with use of the internet connection. The figure below visually shows how the
computers and devices are connected.
Figure 18. Laboratory network diagram.
42
Windows XP Computer for Vacuum System, Dye Laser, and Mass Spectrometer
The Windows XP computer connected to the cRIO is used to modify the code that
runs as a real time application on the cRIO. Once the code is modified, a real time
application is built and automatically run at the next reset of the cRIO. The cRIO’s real
time application does not have a front panel, so a connected computer is not required
once the application is built. The real time application runs as long as the cRIO has
power. The cRIO’s real time application uses “Real Time Panel.vi” as the main VI, and
uses the following VIs as subVIs: “Push Button Panel.vi”, “ControlsSubVI.vi”, “Status
Text.vi”, “Alarm Conditions.vi”, and “LCD Display Loop.vi”.
“Real Time Panel.vi.” “Real Time Panel.vi” is the main VI in the real time
application and monitors the several pressure gauges and thermocouples. These values
are stored in variables that can be accessed from any VI on the network. “Real Time
Panel.vi” also monitors the alarm conditions of the vacuum system using “Alarm
Conditions.vi” as a subVI.
“Alarm Conditions.vi” “Alarm Conditions.vi” will activate the piezo alarm and
make the appropriate change to the system if the laser power supply’s temperature
reaches 104o F, the diffusion pump’s water line temperature reaches 120o F, the diffusion
pump temperature reaches 665o F, the diffusion pump’s water flow decreases below 0.25
gal/min when the diffusion pump is in use, the chamber pressure is greater than 975 Torr,
the smoke detector is triggered, or the gate valve is neither open or closed.
“LCD Display Loop.vi” “Real Time Panel.vi” uses “LCD Display Loop.vi” as a
subVI to display values on the four LCD displays. The pressure of the chamber is
displayed on the upper left LCD display. The other three LCD displays have three43
position switches on the vacuum cart to display temperatures from all the of the
thermocouples in use, and the max temperature reached of the laser’s power supply. The
max temperature reached of the power supply is left visible once the fan is turned off
after use and resets once the fan is turned on again for the next use. The lower right
three-position switch is also used to turn on and off the extra installed fan for the ND:
YAG power supply.
Figure 19. Vacuum cart push panel.
The left figure shows the system at a stable state with no valves open or pumps running. The “Panel Control” push button is also off
so no changes to the system could be made with the other buttons. The right figure shows the vent valve open and the radiation cover
open. The “Panel Control” push button is also on allowing control of the system from the push panel.
“Push Button Panel.vi, ControlsSubVI.vi, Status Text.vi” “Real Time Panel.vi”
also monitors the “Panel Control” push button on the push button panel of the vacuum
system. When the button is held and an automated cycle is not in progress, the button
will light up and the push buttons can be used because of a notifier being sent to run
“Push Button Panel.vi”. When a push button is pressed, “ControlsSubVI.vi” checks to
see if the change is safe for the system, makes the change, lights up the appropriate
button on the push panel, and writes the change made to a string using “Status Text.vi”.
Any change made from any VI that can make a change uses “ControlsSubVI.vi”. This
44
results in the push panel displaying the state of the chamber at all times and all changes
are recorded to a string with a time stamp. The string of all system changes from any VI
can be accessed using “Vacuum System Contol.vi” described next. When the push panel
is finished being used, holding the “Panel Control” push button sends a notifier to end
“Push Button Panel.vi” and turns the light of the “Panel Control” push button off.
Buttons cannot be pressed to change the system without turning the “Panel Control” push
button on.
“Vacuum System Contol.vi, Cycle Control.vi, and Rad Cover Cycle.vi” “Vacuum
System Contol.vi” is located in the same LabVIEW project on the same computer the
cRIO is connected to. “Vacuum System Contol.vi” is used to control the system from a
computer and use two available automated tasks. “Vacuum System Contol.vi” has three
tabs available on the front panel: “Controls”, “Automated Task”, and “Maintenance”.
45
Figure 20. Front panel of Vacuum System Control VI on “Controls” tab.
46
The “Controls” tab shown in Figure 20 has buttons and indicators controlling the
vacuum system and displaying the system’s state. A “MIV#2 Filler” button is available
to back-fill the chamber and have Manifold Isolation Valve #2 automatically close when
the desired pressure is reached. Any button pressed runs through “ControlsSubVI.vi” as
described before. A “Status/Alarms” field on this tab shows the string of all system
changes described before. When the “Reset Status/Alarm Report” button is pressed, the
entire status string prints to file and becomes available on the desktop. The status string
is flushed once printed to file. The second tab is the “Automated Task” tab and is shown
in Figure 21.
47
Figure 21. Front panel of Vacuum System Control VI on “Automated Task” tab.
48
Indicators for the system’s state are available for viewing and “Cycle Control.vi”
and “Rad Cover Cycle.vi” are used as subVIs. “Cycle Control.vi” performs a userdefined number of “vacuum-cycle gas” cycles, before the final “vacuum-sample gas”
cycle for the sample gas mixture. A “vacuum-cycle gas” cycle consists of evacuating the
chamber below 0.15 Torr using Roughing Pump #2. The “Use Diff Pump?(*)” button
decides whether to begin using the diffusion pump or continue using Roughing Pump #2.
The appropriate method continues to evacuate the chamber for a user-defined length of
time. Once that time is reached, all valves are closed and the chamber is back-filled with
the cycle gas, or sample gas mixture for “vacuum-cycle gas” cycles, to the pressure
defined by the user. Radio buttons allow selection of the shutoff valve that is used for the
cycle and sample gas. Using the appropriate fields, the final sample can be a mixture of
two gases for customization of the chamber’s atmosphere. If the final sample gas is a
100% mixture and comes from a shutoff valve other than the cycle gas’s shutoff valve, an
extra “vacuum-sample gas” cycle is performed to assure the final sample gas does not
have any cycle gas from the manifold line from the previous “vacuum-cycle gas” cycles.
To run consecutive experiments, the “Cycle” tabs allow customization of complete
“vacuum-cycle gas” and “vacuum-sample gas” cycles with different final sample
mixtures followed by a radiation cover cycle. To enable this feature, the “Enable for
Multiple Concentration Cycles?” button has to be true. When this feature is on, a field
for “How Many Different Concentration Cycles?” and “Hours to Wait for Ringdown to
Run” become visible. The “Hours to Wait for Ringdown to Run” field is used for a wait
timer that allows the CRDS τ0 measurement to be made, then the following radiation
cover cycle to run. After the wait timer is complete, the next round of “vacuum-cycle
49
gas” and “vacuum-sample gas” cycles begin. Any change to the vacuum chamber system
by “Cycle Control.vi” uses “ControlsSubVI.vi” as mentioned before. An automated
vacuum cycle cannot start with the “GO” button if the system is not in the ready state
indicated by the “*”, used when wanting to use the diffusion pump, and “^”, used when
wanting just Roughing Pump #2, next to the system’s indicators. “Rad Cover Cycle.vi”
is used to automate cycles of covering and uncovering the radiation source cover. The
subVI cycles the linear feedthrough cover a user-defined number of cycles for a userdefined duration at each step (x). A cycle starts at the closed position and remains closed
for x minutes. The cover is then opened and remains open for x minutes. When the
cover closes, one cycle is complete. A button is available to choose whether this cycle
starts opened or closed and alternates accordingly. If the “Vacuum System Contol.vi” is
running and on the “Automated Tasks” tab, the ringdown software has an option to start
the radiation cover cycle with the current parameters entered using the ringdown
software’s “Timer to Release Tau0” button. The “Timer to Release Tau0” button is
described in more detail in the upcoming “Ringdown Time.vi” section. “Rad Cover
Cycle.vi” uses “ControlsSubVI.vi” to change the position of the radiation cover.
The final tab available in “Vacuum System Contol.vi” is the “Maintenance” tab.
The “Maintenance” tab has indicators displaying the start cycles and run hours of both
roughing pumps, the diffusion pump, and the ND: YAG’s power supply fan. The
counters were started when the oil was last changed in the three pumps. A button is
available to reset these counters after the next oil change without affecting the ND:
YAG’s power supply fan’s statistics and vice versa.
50
“Dye Laser Control.vi” “Dye Laser Control.vi” and “UGA.vi” are located in the
same project as “Vacuum System Contol.vi”. “Dye Laser Control.vi” was made to create
a locking burst scan that would use the LambdaLok wavelength measurement feature of
the dye laser, because the factory software did not have this feature.
51
Figure 22. Front panel of Dye Laser Control VI on “Lock” tab.
A burst scan uses a user-defined start and finish wavelength and steps with a userdefined wavelength increment. A button is available to select whether to enter a time in
the “Time at Step (s)” field to wait at each step or wait for a boolean change as a result of
52
the “Timer to Release Tau0 then Start Rad Cover Cycle” button in “Ringdown Time.vi”
(see page 62). The boolean sent from this result is described in more detail in the
upcoming “Ringdown Time.vi” section. “Dye Laser Control.vi” can also lock to a
wavelength using LambdaLok, perform a constant speed scan, and perform a burst scan
without LambdaLok. When the doubling crystals are being used, the “Using Doubling
Crystal?” button divides all outputted wavelengths by two.
53
Figure 23. Front panel of Dye Laser Control VI on “Sirah Control” tab.
“UGA.vi” “UGA.vi” was made to automate the UGA. “UGA.vi” uses the
factory’s automated start and stop sequence for controlling the UGA. Buttons are
available to turn the filament on and off along with opening and closing the sample valve.
54
The UGA can perform analog scans, histogram scans, and single mass measurements.
During an analog scan, the mass spectrometer is stepped at fixed mass increments
through a user-defined mass-range. The advantage of an analog scan is the resolution of
the scan. A histogram scan steps at integer mass increments through a user-defined
mass-range, recording only data points for each integer mass. Two advantages of
histogram scans are faster scan rates than analog scans, and a reduced amount of data
exchanged during the scan. Histogram scans can be used to see the time evolution of
different masses. Single mass measurements can monitor a single mass or be used to
monitor several different masses to display their partial pressures as a function of time.
55
Figure 24. Front panel of UGA VI.
Windows XP Computer for Data Collection
The Windows XP computer controlling the PXI uses two VIs and is the computer
responsible for outputting the data to text files. “PSP Variables for LV9XP.vi” and
“Ringdown Time.vi” are the two VIs used from the LabVIEW project and are shown in
Figure 25.
56
Figure 25. Front panel of Ringdown VI and PSP Variables VI.
The “Small Front Panel Mode” button is on, allowing the project and PSP Variables VI to be seen.
“PSP Variables for LV9XP.vi” “PSP Variables for LV9XP.vi” is responsible for
reading the vacuum system’s temperatures, pressures, system state, and the dye laser’s
variables using the SVE and NI-PSP to update local variables that are used by
“Ringdown Time.vi” for data output and syncing of several tasks with the other Windows
XP computer. “PSP Variables for LV9XP.vi” was created in case an error occurred
transferring data between the two computers; “Ringdown Time.vi” would continue to run
while the issue was addressed. “Ringdown Time.vi” cannot start if “PSP Variables for
LV9XP.vi” is not running and updating all variables. If “Ringdown Time.vi” is started
57
without “PSP Variables for LV9XP.vi” running and ready, a dialog box will notify the
user. This feature was added so “PSP Variables for LV9XP.vi” is remembered.
“Ringdown Time.vi” “Ringdown Time.vi” is used to aid in evaluating cavity
alignment as well as measure and record all data.
Figure 26. Front Panel of Ringdown Time VI.
The “Small Front Panel Mode” button is off, allowing the exponential fit and solved concentration to be seen.
This VI graphically displays the PMT’s waveform, the exponential fits, a plot of
ringdown times calculated using the cavity length, and a plot of concentration once a τ0
value is calculated. A field is available to enter the cross section of the atom or molecule
that is expected to be measured (σ), and the number of laser pulses to be averaged. A
slider is available to select the portion of the waveform for exponential fitting. Indicators
58
display the ringdown time, τ0, standard deviation of τ0, and reflectivity. The “Other” tab
has an input field for the length of the cavity (d), and indicators of several exponential fit
values that are not regularly used. “Ringdown Time.vi” has an align mode when the
“Align Mode” button and the “Block Average” button are selected and is shown in Figure
27.
Figure 27. Front panel of Ringdown VI in Align Mode.
An extra blue slider becomes available where the user can define two separate
regions of the waveform to have two exponential fits calculated from a block average of a
user-defined number of waveforms. The ringdown times are then calculated and a ratio
of the two times are used for alignment. With block average running, the cavity can be
59
finely adjusted until the ratio of the two ringdown times is close to 1. With a ratio of 1,
the cavity is excited in a manner that yields a high quality single exponential decay.
When alignment is complete, the “Align Mode” and “Block Average” button are toggled
off, and the blue slider and ratio indicator disappear. With the “Block Average” button
off, a running average is used. The number of waveforms to be averaged uses the same
field that block average used. When the small ND: YAG 20 Hz laser was used, 20
averages were used for one second data points. When the large ND: YAG 30 Hz laser
was used, 30 averages were used for one second data points. A “Mark” button can output
the number 1000 to the Mark column of the experimental file for use in analysis. A
“Reset Statistics?” button is available so the user can adjust sliders, change other
parameters, and then be able to reset the τ0 average and standard deviation without
restarting the VI. Two options are available for calculating τ0. When on, the “Tau0”
button will calculate an average until the button is released. When released, the average
will be used as τ0 and every data point thereafter is used to calculate the concentration of
the absorber using the provided cross section. The second τ0 option is the “Timer to
Release Tau0” button. When on, the “Tau0” button is disabled and a field to enter “Time
to Wait in Minutes” is visible. When the ringdown VI’s clock reaches that time, the
average is used as τ0 to begin calculating concentration. A boolean variable is also
flipped, telling the “Rad Cover Cycle.vi” to begin on the other Windows XP computer as
briefly described before. This feature allows one to automate several experimental runs
consecutively. When all parameters satisfy the user, the “Write Data to File” button will
do the function of the “Reset Statistics?” button, reset the VI’s clock, and generate four
paths into a folder using the date and time stamp. The “experimental” file, or abbreviated
60
“exp” in the file name, updates after every averaged waveform and contains columns of
the VI’s clock, ringdown time calculated from the exponential fit, standard deviation of
the ringdown time, r2 value of the exponential fit, amplitude of the PMT’s averaged
waveform, running average of ringdown times (τ0), “Mark” value used, upper capacitance
manometer pressure, lower capacitance manometer pressure, ion gauge pressure,
LambdaLok lock feature’s set wavelength, LambadLok wavelength value, LambdaLok
wavelength deviation, and thermocouple readings for the laser power supply, room
temperature, inner chamber wall, diffusion pump, and chamber atmosphere. There are
several values that can be used for the “Mark” function. The number 750 is outputted
when reset statistics is used, 1000 is used when the physical “Mark” button is pressed,
1250 is used when the radiation source cover closes, 1500 is used when the radiation
source cover opens, 1750 is used when an automated cycle closes Manifold Isolation
Valve # 2, and 2000 is used when an automated cycle opens Manifold Isolation Valve #
2. The experimental file outputs data to text as long as the “Write Data to File” button is
on. A “comment” file is also created to log any comment entered into the comment field
along with the value of the VI’s clock. The third file path created is a waveform file.
When the “Save Waveform” button is pressed, two columns of time and voltage print
once the waveform has averaged. The final file path created is a “concentration” file, or
abbreviated “conc” in the file name. Once τ0 is calculated and concentration can be
measured, this file updates. The concentration file updates after every averaged
waveform and outputs columns of VI’s clock, ringdown time calculated from the
exponential fit, number density of absorber, part per billion of absorber using
Loschmidt’s number, part per billion of absorber using Loschmidt’s number scaled to the
61
chamber’s real time pressure, part per billion of absorber using the number density
calculated from the ideal gas law using the chamber’s real time pressure and temperature,
and the value of any “Mark” that is used. “Ringdown Time.vi” has one more feature, an
“Enable Ringdown to Step Scan?” button, for use when scanning known concentrations
of an absorber and solving for the absorber’s cross section. The “Enable Ringdown to
Step Scan?” button can be selected so a boolean variable will flip after a user defined
number of data points have been recorded, telling the dye laser to move to the next
wavelength. This feature was used to solve for the cross section of sulfur dioxide (SO2)
in this thesis.
Experiment Automation
For the purpose of clarifying VI relationships, the computer connected to the
cRIO will be referred to as computer A, and the computer used for ringdown
measurements will be referred to as computer B. Several experimental runs can be
automated consecutively with the “Enable for Multiple Concentration Cycles?” button on
from the “Automated Tasks” tab of “Vacuum System Contol.vi” (see page 49) on
computer A. When “Vacuum System Contol.vi” completes the vacuum chamber’s cycles
and fills the chamber with the test sample, it will programmatically toggle the “Write
Data to File” button in “Ringdown Time.vi” (see page 60) from computer B off then on,
resetting the clock to zero and generating new file paths. When the “Ringdown Time.vi”
clock reaches the “Time to Wait in Minutes” value from the “Timer to Release Tau0”
button being on, τ0 is ready, concentration data begins logging, and the boolean variable
is flipped telling the “Rad Cover Cycle.vi” (see page 50) on computer A to begin the
radiation cover cycle. Once the radiation cover cycle is complete, “Vacuum System
62
Contol.vi” toggles the “Write Data to File” button in “Ringdown Time.vi” from computer
B. This entire process is then repeated for the next sample gas. Automated toggling of
the “Write Data to File” button generates two sets of files for each sample gas. One set
of files has ringdown data and vacuum system data during the vacuum automation cycle,
and the second set of files contains all data on the sample gas starting at the beginning of
the τ0 measurement until the end of the radiation cover cycle. The file sets are named
with a time stamp and capital letter at the end of the filename string corresponding to the
tab used by “Vacuum System Contol.vi” to set the sample gas’s gas mixture.
Windows 7 Computer for Remote Login and Email Notifications
The Windows 7 computer is used to permit remote access to the other two
computers using VNC Viewer and to send notification emails to the user using “Alarm
for Win7.vi”.
“Alarm for Win7.vi” “Alarm for Win7.vi” monitors the boolean values of “Dye
Scan Complete”, “Vacuum Cycle Complete”, “Rad Cover Cycle Complete”, “Ringdown
Error”, “Lambda Lock Error”, “Vacuum Cycle Chamber Fill Error”, and any alarm
condition in the “Alarm Conditions.vi”. When any of these boolean values become true,
an email describing the scenario is sent to the email recipients. The email recipients are
entered on the front panel of “Alarm for Win7.vi” shown in Figure 28.
63
Figure 28. Front panel of Alarm VI.
64
CHAPTER VIII – DATA COLLECTION AND RESULTS
Before sensitive ozone measurements were made, the quantitative accuracy of the
system was examined along with characterization of the ringdown measurement stability
throughout pressure cycles as well as during static conditions. Because the optics table
was not rigidly mounted to the vacuum system, any vibrations experienced by the
vacuum system pumps or UGA would not be experienced by the laser systems. To
confirm that the system was unaffected by these vibrations and was capable of making
stable ringdown measurements, pressure cycles between low pressure (~150 mTorr) and
760 Torr of Ultra High Purity (UHP) N2 were performed. Two 266 nm centerwavelength CRD mirrors were used to make 266 nm and 283 nm CRDS measurements.
The laser output was directed from the optics table into the CRD cavity using the mirrors
described in Table 2 of Chapter VI. The 266 nm and 283 nm experiments were
performed separately. The 266 nm data was used to measure Rayleigh scattering losses
due to N2 while both the 266 nm and 283 nm data was used to validate measurements
using SO2.
The Rayleigh scattering cross section for N2 was calculated at 266 nm to be 1.07 x
10-25 cm2/molecule 39. The predicted shift in ringdown time for each cycle of low
pressure to 760 Torr of UHP N2 was calculated from equation (4.12) from Chapter IV
for 266nm:
𝑁 = 1⁄𝑐𝜎(𝜆) (1⁄𝜏 − 1⁄𝜏0 )
𝜏=
𝜏0
⁄(1 + 𝜏 𝑁𝑐𝜎)
0
U Griesmann and J.H. Burnett, “Refractivity of Nitrogen Gas in the Vacuum Ultraviolet,” Opt. Lett. 24
(1999): 1699–1701; D. Jackson, Classical Electrodynamics (John Wiley, 1975).
39
65
𝛥𝜏 = 𝜏0 − 𝜏 = 𝜏0 −
𝜏0
⁄(1 + 𝜏 𝑁𝑐𝜎) .
0
(8.1)
Using the calculated number density (𝑁), calculated cross section (𝜎), and the average
ringdown time at low pressure as the reference ringdown time (𝜏0 ), the predicted shift in
ringdown time for each cycle was determined. The one hour τ0 measurement had a
standard deviation of 15.9 ns. The shift in ringdown times during pressure cycles was
consistent with the predicted values as shown below in Table 4:
Table 4
Predicted Shift of Average Ringdown Values from Vacuum-N2 Cycles for Rayleigh
Scattering
Wavelength
(nm)
Cycle
Number
Ringdown
Time at
~150
mTorr
Ringdown
Time at
~760
mTorr
UHP N2
(ns)
266
1
2
625
625
587
586
Change in
Ringdown
Time Δτ
(ns)
38
39
Predicted
Shift of
Ringdown
Time Due
to
Rayleigh
Scattering
Δτ (ns)
30
30
Percent
Error
(%)
27
30
Before every measurement was made, two automated cycles between low pressure and 760 Torr of N2 were cycled to clear the
chamber before the sample gas for the radiation exposure measurement was introduced. These two N 2 cycles can be used to check the
Rayleigh scattering shifts for any measurement made in this thesis and is available as separate files for later analysis.
This data validates that the data acquisition and analysis are being performed
correctly, and the system is stable throughout pressure cycles as well as static conditions.
Figure 29 below shows the ringdown measurements of the Rayleigh scattering of N2 at
266 nm.
66
Figure 29. Ringdown times (ns) vs time (hrs) of N2 pressure cycles at 266 nm.
Ringdown times measured at 266 nm wavelength at low pressure (150 mTorr, 1st and 3rd hours) and at atmospheric pressure (760
Torr, 2nd and 4th hours) using UHP N2. The red line is the original data acquired using an average of 20 laser pulses, while the black
line is an average of 50 of the red data points.
An alternate depiction of the same data is shown in Table 5 below, however, the
measured ringdown times and known Rayleigh scattering cross section were used to
calculate the number density of N2 molecules near atmospheric pressure. For species
with unknown concentrations, such as the radiation-induced species to be measured, this
was the form of analysis used. Table 5 shows the calculated shift of the number densities
and the experimental measurement of the shift of number densities:
67
Table 5
Predicted Number Density from Vacuum-N2 Cycles for Rayleigh Scattering
Wavelength
(nm)
Cycle
Number
Measured Number
Density
(molecules/m3)
266
1
2
2.49 x 1025
2.49 x 1025
Predicted Number
Density Due to
Rayleigh Scattering
(ns)
3.23 x 1025
3.32 x 1025
Percent
Error (%)
23
25
Because these values differ substantially, a clearer evaluation of the accuracy of
the system for absorption spectroscopy measurements was performed using sulfur
dioxide (SO2).
For further validation of the CRDS system at 266 nm and the dye laser’s scanning
ability, a known concentration of SO2 (50.13ppm) balanced in N2 was purchased and
used to experimentally validate the system. The SO2 was prepared and experimentally
validated by the gas supplier. SO2 has a unique absorption cross section and is shown in
Figure 30:
68
Figure 30. SO2 absorption cross section from 238 nm – 325 nm.
This figure has been reproduced with permission (See Appendix A) 40.
An SO2 absorption measurement was made at 266 nm using the small ND: YAG
laser for validation of the system accuracy. The system was cycled three times with UHP
N2 before two cycles of 100 Torr of SO2 mixed in N2. A τ0 measurement of 638.91 ns
with a 19.5 ns standard deviation was made at 100.09 Torr of UHP N2 at 295.73 K over a
thirty-minute period. A τ measurement of 287.95 ns was made on 97.06 Torr of 50.13
ppm SO2 balanced in N2 over a thirty-minute period. Solving for the cross section of SO2
at 266 nm, our measurement was 4.08 x 10-19 cm2/molecule. This measurement has a
Keller-Rudek et al., “The MPI-Mainz UV/VIS Spectral Atlas of Gaseous Molecules of Atmospheric
Interest”; J Rufus et al., “High-Resolution Photoabsorption Cross Section Measurements of SO2, 2: 220 to
325 Nm at 295 K,” J. Geophys. Research - Planets 108 (2003).
40
69
10% percent error agreement with 4.52 x 10-19 cm2/molecule measurement made by
Vandaele at 296K at 266.02 nm 41.
An SO2 measurement was also made from 282 nm – 284 nm using the Sirah dye
laser pumped by the 532 nm output from the Spectra Physics ND: YAG. Rhodamine 6G
dye was used at Sirah’s recommended concentration. Because the concentration and
testing pressure of the SO2 mix was known, a scan using LambdaLok wavelength
monitoring from 282 nm – 284 nm at 0.1 nm per step was performed to solve for the
unique cross section of SO2. The cross section was solved for by scanning a chamber of
only N2 at a certain pressure and using that scan as τ0. The same set of scan parameters
was performed on a chamber of known SO2 concentration in N2 (τ) at the same pressure
of the first scan. The data points during the five minutes at each scan step were averaged
to one data point. Using these data points, the unique cross section from 282 nm – 284
nm was determined. The following four measurements were made: three measurements
conducted near a total pressure of 100 Torr while using different SO2 concentrations, and
one measurement conducted at 28 Torr using the calibrated concentration. In all cases,
the experimentally measured absorption cross sections agreed well with five literature
reports 42. Figure 31 below shows the SO2 absorption cross section from the five
literature reports (solid lines) and our four experimental measurements (dashed lines):
A.C. Vandaele et al., “SO2 Absorption Cross Section Measurement in the UV Using a Fourier Transform
Spectrometer,” J. Geophys. Res. 99 (1994): 25599–605.
42
B.S. Olive, “Absorption Spectra of Benzene, Toluene, and Sulfur Dioxide” (Florence, AL: University of
North Alabama, 2015); B.S. Olive, “Absorption Spectrum of Sulfur Dioxide” (Florence, AL: University of
North Alabama, 2007); A.C. Vandaele and S Fally, “Fourier Transform Measurements of SO2 Absorption
Cross Sections: II. Temperature Dependence in the 29 000-44 000 Cm-1 (227-345 Nm) Region,” J. Quant.
Spectrosc. Rad. Transfer 110, no. 18 (2009): 2115–26; S.M. Ahmed and Vijay Kumar, “Quantitative
Photoabsorption and Flourescence Spectroscopy of SO2 at 188–231 and 278.7–320 Nm,” J. of Quant
Spectrosc. Rad. Transfer 47, no. 5 (1992): 359–73; K Bogumil et al., “Measurements of Molecular
Absorption Spectra with the SCIAMACHY Pre-Flight Model: Instrument Characterization and Reference
41
70
Figure 31. Experimental SO2 absorption cross section compared to literature reports.
SO2 absorption cross section from the five literature reports (solid lines) and four experimental measurements (dashed lines). Standard
deviations of each run can be found in their respective files.
In all cases, the measured cross section agreed well with the literature reports, and
validates that these CRDS measurements are quantitatively accurate and the dye laser and
LambdaLok scan work as planned. With the SO2 measurements of both laser systems in
good agreement with measured values from other literature reports, the system is
believed to be capable of making sensitive and accurate measurements.
Data for Atmospheric Remote-Sensing in the 230-2380 Nm Region",” J. of Photochem. and Photobio 157,
no. 2–3 (2003): 167–84.
71
Ozone (O3) is a well-known product of air radiolysis and is a strong absorber at
266 nm with a cross section of 9.44 x 10-18 cm2/molecule 43. The small 20 Hz ND: YAG
laser has the required crystals for the 4th harmonic 266 nm output, unlike the larger ND:
YAG. Multiple runs in 80%/20% N2/O2 mixtures were performed using a mixture
verified by the gas supplier. The runs were performed at the target pressure of 750 Torr.
Before each fill of the sample gas, the chamber was evacuated and purged twice with
UHP N2. All cycling is controlled and automated by the LabVIEW programs for
reproducibility. Multiple runs were performed that included one, two, and three-hour τ0
measurements with the radiation sources covered, followed by five radiation cover cycles
with ten or sixty minutes at each position. Runs were also performed with one-hour τ0
measurements and five ten-minute radiation cover cycles in 0%, 10%, and 20% O2
concentrations. Two 5 mCi Po210 alpha sources were covered and uncovered during the
radiation cover cycles.
Ringdown time and ozone concentration plots are available to show the rapid
production of ozone immediately following the exposure of the radiation. Note that the
concentrations in all figures below are average concentrations over the length of the CRD
cavity. In actuality, the alpha particles only impact a fraction of the cavity length as
shown by the penetration calculations in Table 1 of Chapter II. Estimating the diameter
of the alpha excited region to be 8 cm and given a cavity length of 43.6 cm, the actual
concentration of ozone above the source was estimated to be a factor of 43.6/8, or a factor
of approximately five higher than the y-axis label in the concentration figures.
Bogumil et al., “Measurements of Molecular Absorption Spectra with the SCIAMACHY Pre-Flight
Model: Instrument Characterization and Reference Data for Atmospheric Remote-Sensing in the 230-2380
Nm Region".”
43
72
Figure 32 below represents the ozone concentration calculated by runs performed
on two different dates with a one-hour τ0 measurement, and five ten-minute radiation
cover cycles. After the one-hour τ0 measurement was averaged (11 ns standard deviation
average), the radiation cover cycle was started and the τ0 average was used along with the
number density of the chamber to calculate the concentration of ozone using the cross
section mentioned above.
Figure 32. Ozone concentrations after one-hour τ0, five ten-minute radiation cycles with
20% O2 in N2.
Eight runs (different colors) were performed above with two 5 mCi 210Po alpha sources at a target pressure of 750 Torr. The sources
were shielded for one hour, and then repetitively shielded/exposed for ten minutes for a total of five cycles. The data above shows the
concentration levels of ozone being measured once the radiation source cover cycle was started.
73
As shown in Figure 32, the ozone concentration grows in time during each tenminute period that the radiation source was exposed. Once the radiation source was
shielded, the ozone quickly drops back to near baseline. An interesting effect was
observed whenever the radiation source was exposed again. The ozone concentration
surprisingly did not repeat the previous period’s data. Instead, the ozone level started
growing from a value similar to the one at which the previous period’s concentration
ended. This effect resulted in the ozone concentration getting larger during each cycle of
exposure while falling to near baseline during every shielded cycle. These results suggest
that the radiation caused changes to the atmosphere that aided in the formation of ozone,
but these changes were stable during the shielded cycles.
Figure 32’s ozone concentration data was recorded in real time using the
chamber’s atmospheric state to calculate the number density in the chamber and then
solved for the concentration of ozone due to the shift in ringdown times assuming ozone
was the only new absorber introduced after the τ0 measurement. The recorded ringdown
times during the one-hour τ0 measurement and the radiation cover cycles show another
interesting effect in Figure 33 below.
74
Figure 33. Ringdown times of one-hour τ0, five ten-minute radiation cycles with 20% O2
in N2.
These five measurements were included in Figure 32. Five runs (different colors) were performed above with two 5 mCi 210Po alpha
sources at a target pressure of 750 Torr. The sources were shielded for one hour, and then repetitively shielded/exposed for ten
minutes for a total of five cycles.
A slow decrease in ringdown time was evident while the radiation source was
shielded which may correlate to an optical loss due to some absorber. To extend the
above hypothesis of a stable ozone catalyst, a small volume of air is present under the
aluminum radiation source shield that could possibly result in gas-phase species being
created that are leaking into the chamber between the aluminum shield and source holder.
The shield was not designed to be airtight. The species responsible for the small optical
loss is not likely to be ozone because ozone disappeared rapidly when the radiation
75
source was shielded. Ozone is also very reactive with the materials of the vacuum
chamber.
To further investigate the effect that occurs during the τ0 measurement, longer τ0
measurements were performed before starting the same radiation source cycle. Three
runs were performed with two-hour τ0 measurements, and several runs were performed
with three-hour τ0 measurements. Following the τ0 measurements, the same radiation
source cover cycles of five ten-minute radiation cycles were performed. Figure 34 and
Figure 35 show the continuing decrease in ringdown time during the two and three-hour
τ0 measurements. In Figure 35, the light blue line represents the same three-hour τ0
measurement parameters being performed with no radiation sources in the chamber to
show the drift in ringdown is a radiation-induced effect, and not a result of a nonradiation related process.
76
Figure 34. Ringdown times of two-hour τ0, five ten-minute radiation cycles with 20% O2
in N2.
Three runs (different colors) were performed above with two 5 mCi 210Po alpha sources at a target pressure of 750 Torr. The sources
were shielded for two hours, and then repetitively shielded/exposed for ten minutes for a total of five cycles. The average standard
deviation of the three runs was 10 ns.
77
Figure 35. Ringdown times of three-hour τ0, five ten-minute radiation cycles with 20%
O2 in N2.
Five runs (different colors) were performed over two days with two 5 mCi 210Po alpha sources at a target pressure of 750 Torr. The
sources were shielded for three hours, and then repetitively exposed/shielded for ten minutes for a total of five cycles (to line up with
later measurements). The light blue line represents the same run being performed with no radiation sources in the chamber. The light
blue run was performed with a different alignment than the other measurements. The different alignment is the reason for an initial
difference in ringdown time between the light blue line and the other lines of approximately 10 ns. The average standard deviation of
the three runs was 10 ns.
The concentration of ozone calculated using the one, two, and three-hour τ0
measurements is plotted on one graph below as a function of time since the chamber was
filled. Figure 36 shows the ozone levels starting at higher levels due to the longer τ0
measurements. Once the sample gas was introduced into the chamber, the radiation
78
source was shielded but still able to cause a slow change in the atmosphere that catalyzed
the formation of ozone.
Figure 36. Ozone concentration from one, two, and three-hour τ0 measurements as a
function of time from when the chamber was filled.
The red lines represent the one-hour τ0 runs, the black lines represent the two-hour τ0 runs, and the pink lines represent the three-hour
τ0 runs. All runs were performed with two 5 mCi 210Po alpha sources at a target pressure of 750 Torr. The sources were shielded for
during the respective τ0 measurement, and then repetitively shielded/exposed for ten minutes for a total of five cycles. The data above
shows the concentration levels of ozone being measured once the radiation source cover cycle was started.
Measurements were also performed with five one-hour radiation source cover
cycles following two-hour τ0 measurements. Figure 37 below shows a similar drift in
ringdown time during the τ0 measurement. Once the source was exposed, the ozone
quickly grew and continued to grow during the first hour of exposure. Once the source
79
was re-shielded, the ringdown returned to the same level before the first exposure, and
continued this behavior throughout the remaining run suggesting a saturation of the effect
that is causing the drift in ringdown time.
Figure 37. Ringdown times of two-hour τ0, five one-hour radiation cycles with 20% O2 in
N2.
Six runs (different colors) were performed above with two 5 mCi 210Po alpha sources at a target pressure of 750 Torr. The sources
were shielded for two hours, and then repetitively shielded/exposed for one hour for a total of five cycles. The average standard
deviation of the six runs was 11 ns.
The data from Figure 37 was also plotted to show the concentration of ozone
below in Figure 38. The concentration plot shows the growth of ozone during the first
hour of exposure similar to the ten-minute cycles. The concentration for the second to
fifth hour appears to have reached a saturation level.
80
Figure 38. Ozone concentrations after two-hour τ0, five one-hour radiation cycles with
20% O2 in N2.
Six runs (different colors) were performed above with two 5 mCi 210Po alpha sources at a target pressure of 750 Torr. The sources
were shielded for two hours, and then repetitively shielded/exposed for one hour for a total of five cycles. The data above shows the
concentration levels of ozone being measured once the radiation source cover cycle was started.
The concentrations of the ten-minute radiation cover cycles following the threehour τ0 measurements were plotted together with the concentrations of the one-hour
radiation cover cycles following the two-hour τ0 measurements. These plots were
combined because the radiation sources were exposed for the first time after being
shielded for the same length of time once the sample gas was introduced. As expected,
the initial level of ozone was reached by both the ten-minute and one-hour radiation
cover cycles. The pink line was also plotted and represents a run of ten-minute radiation
81
cycles following a three-hour τ0 measurements with no sources in the chamber. The pink
line verifies that the motion of the linear feedthrough that shields and exposes the
radiation source is not a cause for a shift in ringdown times.
Figure 39. Ozone concentrations of ten-minute and one-hour radiation cover cycle
following three hours of shielding with 20% O2 in N2.
The red lines represent the three-hour τ0 measurement with ten-minute radiation cover cycles starting at exposed. The black lines
represent the two-hour τ0 measurement with one-hour radiation cover cycles starting at shielded. The pink lines represent the threehour τ0 measurement with ten-minute radiation cover cycles of no radiation sources. All other runs were performed with two 5 mCi
210
Po alpha sources at a target pressure of 750 Torr. The sources were shielded for during the respective τ0 measurement, and then
repetitively shielded/exposed for ten minutes for a total of five cycles. The data above shows the concentration levels of ozone being
measured once the radiation source cover cycle was started.
The initial concentration of oxygen was varied before five ten-minute radiation
cycles were performed. Several runs were performed with 20%/80% O2/N2, 10%/90%
82
O2/N2, and 0%/100% O2/N2 as the initial sample gas. No distinct difference between ten
and twenty percent O2 were noticed, but as expected, no ozone was produced in the zero
percent O2 run. Figure 40 shows these results below.
Figure 40. Ozone concentrations after one-hour τ0, five ten-minute radiation cycles with
zero, ten, and twenty percent oxygen balanced in nitrogen.
The black line represents the 0% O2 run, the gray lines represent the 10% O2 runs, and the red lines represent the 20% O2 runs. All
runs were performed with two 5 mCi 210Po alpha sources at a target pressure of 750 Torr. The sources were shielded for one hour, and
then repetitively shielded/exposed for ten minutes for a total of five cycles. The data above shows the concentration levels of ozone
being measured once the radiation source cover cycle was started. The standard deviation values can be found in the respective file.
Interestingly, higher ozone concentrations were measured for two of the 10% O2
mixes, and the 10% O2 mixes never produced an ozone concentration lower than the 20%
83
O2 mixes. This effect could be evidence of N2 being a more efficient third body partner
when the required amount of O2 is still available for the formation of ozone.
Table 6
Radiation Source Manufactured-Date and Date Experiments Were Performed
Dates the Experiments in
the Figures Were
Performed
Radiation Source
Manufactured Date
Figure 32
May 18, 2016 and May 23,
2016
April 2016
Figure 34
May 26, 2016
April 2016
Figure 35
May 27, 2016 and May 28,
2016
April 2016
Figure 37
May 20, 2016
April 2016
Figure 40
November 15, 2015 (0%
O2), May 18, 2016 (20%
O2), and May 19&23, 2016
(10% O2)
October 2015 (0% O2),
April 2016 (others)
Many measurements were made for various scenarios that all showed similar
behaviors. The rapid production of ozone was apparent in every measurement. The data
also showed effects of a slow decrease in ringdown time that was evident while the
radiation source was shielded. The data also showed the ozone concentration getting
larger during each cycle of exposure while falling to baseline during every shielded cycle.
This effect was also shown to saturate over time. These results suggest that the radiation
84
was causing changes to the atmosphere that were aiding in the formation of ozone, but
these changes were stable during the shielded cycles. A comparison of our ozone
measurements with an early-version Cantera kinetics calculation show a similar behavior.
This comparison along with future work and a conclusion will follow in Chapter IX.
85
CHAPTER IX – CONCLUSIONS AND FUTURE WORK
As recently stated, the rapid production of ozone was apparent in every
measurement, but it cannot be said that the only absorbing species produced in the
experiment was ozone. Table 7 shows the predicted concentration of other potential
products of ionizing radiation if the initial shift in ringdown time from Figure 33 of 570
ns to 555 ns was a 100% result of that product. The concentrations were determined
using the product’s absorption cross section at 266 nm.
Table 7
Predicted Concentrations of Other Potential Products of Ionizing Radiation Resulting
from the Observed Shift in Ringdown Time
τ0 = 570 ns
Species
O3
N2O5
NO3
NO2
N2O
τ = 555 ns
Absorption Cross Section
(cm2)
9.44E-1844
1.77E-1945
1.20E-1946
2.40E-2047
5.12E-2448
Concentration
6.52 ppb
348 ppb
513 ppb
2560 ppb
1.20E7 ppb
While achieving the concentration of N2O is unreasonable, the other values are
not necessarily unachievable. O3 is likely responsible for most of the observed effect, but
other species are sure to contribute. As planned under the current project, future work
44
Ibid.
R.A. Graham, “Photochemistry of NO3 and the Kinetics of the N2O5-O3 System” (Ph. D., Univ. of
California, 1975).
46
S.P. Sander, “Temperature Dependence of the NO3 Absorption Spectrum,” J. Phys. Chem. 90 (1986):
4135–42.
47
Bogumil et al., “Measurements of Molecular Absorption Spectra with the SCIAMACHY Pre-Flight
Model: Instrument Characterization and Reference Data for Atmospheric Remote-Sensing in the 230-2380
Nm Region".”
48
D.R. Bates and P.B. Hays, “Atmospheric Nitrous Oxide,” Planet. Space Sci. 15 (1967): 189–97.
45
86
will utilize the dye laser to measure the effects of alpha radiation at wavelengths of other
species’ absorption cross section maxima to create a system of equations for a more
accurate measurement of the effect of alpha radiation. The equipment required for these
measurements is already installed and being used. Using other wavelengths only requires
a different dye solution, mirrors for that particular wavelength, and PMT filters.
The data from this thesis also showed effects of a slow decrease in ringdown time
that was evident while the radiation source was shielded and acquiring a τ0 measurement.
Figure 37 shows two-hour τ0 measurements with an average drift of 5 ns. Table 8 shows
the predicted concentration of products of ionizing radiation if the drift in ringdown time
of the τ0 measurement from 555 ns to 550 ns was a 100% result of that product. As
mentioned in Chapter VIII, ozone is not expected to be the cause of the τ0 drift, but was
included.
87
Table 8
Predicted Concentrations of Products of Ionizing Radiation Resulting from the Observed
Drift in Ringdown Time of τ0 Measurements
τ0 = 555 ns
Species
O3
N2O5
NO3
NO2
N2O
τ = 550 ns
Absorption Cross Section
(cm2)
9.44E-1849
1.77E-1950
1.20E-1951
2.40E-2052
5.12E-2453
Concentration
2.25 ppb
120 ppb
177 ppb
886 ppb
4.15E6 ppb
NO2 has a weak optical absorption at 266 nm, and thus could be responsible for
the slow decrease in ringdown time (or increase in optical absorption) evident in Figure
37.
The data from this thesis also showed one final effect. Figure 32, Figure 36,
Figure 38, Figure 39, and Figure 40 showed the ozone concentration getting larger during
each cycle of radiation exposure while falling to baseline during every shielded cycle.
This effect was also shown to level out over time. These results suggest that the radiation
caused changes to the atmosphere that aided in the formation of ozone, but these changes
were stable during the shielded cycles. As noted in Chapter III, several reaction cycles
exist that involve O3, NO2, NO, N2O, N, and O. The presence of these other species
Bogumil et al., “Measurements of Molecular Absorption Spectra with the SCIAMACHY Pre-Flight
Model: Instrument Characterization and Reference Data for Atmospheric Remote-Sensing in the 230-2380
Nm Region".”
50
Graham, “Photochemistry of NO3 and the Kinetics of the N2O5-O3 System.”
51
Sander, “Temperature Dependence of the NO3 Absorption Spectrum.”
52
Bogumil et al., “Measurements of Molecular Absorption Spectra with the SCIAMACHY Pre-Flight
Model: Instrument Characterization and Reference Data for Atmospheric Remote-Sensing in the 230-2380
Nm Region".”
53
Bates and Hays, “Atmospheric Nitrous Oxide.”
49
88
could aid in the production of ozone. Additionally, several other species are relatively
stable in an atmospheric environment and are compatible with the stainless steel and
aluminum materials of the chamber.
Under the current project, a chemical kinetics Python script that uses the Cantera
toolkit was developed by Patrick Ables. The python script was developed to simulate the
chemical kinetics of the experiment. The script was developed with the ability to provide
continuous introduction of primary reactant species via the mass flow controller process.
The script also has the ability to repetitively turn the introduction of primary reactants on
and off. The mass flow control has been modified to include a square wave time
behavior (i.e. on or off) that can simulate the exposing or shielding of the radiation
source. The results of this Cantera calculation for a number of species are presented in
Figure 41.
89
Figure 41. Early development of Cantera results showing the production and destruction
of ozone during shielded/exposed cycles.
Figure 41 shows the simulation of exposing and shielding the radioactive source similar to the experiment. This figure has been
reproduced with permission (See Appendix A)
Figure 42. Enlarged view of the ozone data from Figure 41.
Figure 42 shows ozone starting at a lower concentration and growing to a saturation level as in the experimental results of Figure 36.
This figure has been reproduced with permission (See Appendix A)
The results show NO2, N2O, and N2O5 remaining relatively stable during the
shielded/exposed cycles. During the shielded cycles, ozone rapidly reacts away. The
90
ozone concentration is also shown to grow in time in a manner very similar to the growth
observed in Figure 36.
Figure 36. Ozone concentration from one, two, and three-hour τ0 measurements as a
function of time from when the chamber was filled.
Figure 41 and Figure 36 each reach ozone saturation between 150 and 200
minutes. The growth in ozone concentration predicted by the Cantera chemical kinetics
calculations, along with the observation of the same effect in experimental data, is
compelling evidence that the radiation physics for primary products and chemical
kinetics calculations are becoming sufficiently accurate to capture the detailed behavior
measured in the experiment. Cavity ringdown measurements for radiation-induced ozone
91
have produced data that is in excellent agreement with the time evolution of ozone
predicted in preliminary chemical kinetics calculations.
As mentioned before, future work will include CRD measurements at a variety of
wavelengths using the dye laser. The vacuum chamber’s capabilities will also allow
other gases (water vapor, trace gases, etc.) to be introduced for further measurements of
radiation-induced effects. A few pieces of equipment could upgrade the experiment to
run on Windows 7 computers only. A new desktop PCI card to connect the PXI would
eliminate the need of the PXI’s Window XP dependent network card. A new wavemeter
would eliminate the need of the dye laser’s Window XP dependent PCI card.
92
APPENDIX A – Permission Emails
93
94
95
REFERENCES
Ackerman, M. “UV-Solar Radiation Related to Mesospheric Processes.” In Mesospheric
Models and Related Experiments, edited by G Fiocco, 149–59. Dordrecht: D.
Reidel Publishing Company, 1971.
Ahmed, S.M., and Vijay Kumar. “Quantitative Photoabsorption and Flourescence
Spectroscopy of SO2 at 188–231 and 278.7–320 Nm.” J. of Quant Spectrosc.
Rad. Transfer 47, no. 5 (1992): 359–73.
Bates, D.R., and P.B. Hays. “Atmospheric Nitrous Oxide.” Planet. Space Sci. 15 (1967):
189–97.
Bogumil, K, J Orphal, T Homann, S Voigt, P Spietz, O.C Fleischmann, A Vogel, et al.
“Measurements of Molecular Absorption Spectra with the SCIAMACHY PreFlight Model: Instrument Characterization and Reference Data for Atmospheric
Remote-Sensing in the 230-2380 Nm Region".” J. of Photochem. and Photobio
157, no. 2–3 (2003): 167–84.
Buchanan, Aubri Capri. “Characterization of Air Fluorescence Induced by Alpha
Radiation.” Master’s Thesis, The University of Southern Mississippi, 2008.
Cole, J., S. Su, R.E. Blakeley, P. Koonath, and A.A. Hecht. “Radiolytic Yield of Ozone
in Air for Low Dose Neutron and X-Ray/gamma-Ray Radiation.” Radiation
Physics and Chemistry 106 (January 2015): 95–98.
doi:10.1016/j.radphyschem.2014.06.008.
Eliasson, B., U. Kogelschatz, and P. Baessler. “Dissociation of O2 in N2/O2 Mixtures.”
Journal of Physics B: Atomic and Molecular Physics 17, no. 22 (1984): L797.
96
Graham, R.A. “Photochemistry of NO3 and the Kinetics of the N2O5-O3 System.” Ph.
D., Univ. of California, 1975.
Griesmann, U, and J.H. Burnett. “Refractivity of Nitrogen Gas in the Vacuum
Ultraviolet.” Opt. Lett. 24 (1999): 1699–1701.
Heicklen, Julian. Atmospheric Chemistry. Academic Press, 1976.
Jackson, D. Classical Electrodynamics. John Wiley, 1975.
Jones, A. Russell. “Radiation-Induced Reactions in the N2-O2-H2O System.” Radiation
Research 10, no. 6 (June 1, 1959): 655–63. doi:10.2307/3570649.
Keller-Rudek, Hannelore, Geert Moortgat, Rolf Sander, and Rudiger Sorensen. “The
MPI-Mainz UV/VIS Spectral Atlas of Gaseous Molecules of Atmospheric
Interest,” n.d. http://www.uv-vis-spectral-atlas-mainz.org/.
Kogelschatz, U., B. Eliasson, and M. Hirth. “Ozone Oeneration from Oxygen and Air:
Discharge Physics and Reaction Mechanisms,” 1988.
L’Annunziata, Michael F. Radioactivity: Introduction and History. Elsevier, 2007.
Lind, S. C. The Chemical Effects of Alpha Particles. American Chemical Society
Monograph Series. New York: The Chemical Catalog Company, 1921.
Lind, S. C., and D. C. Bardwell. “Ozonization and Interaction of Oxygen with Nitrogen
under Alpha Radiation.” Journal of the American Chemical Society 51, no. 9
(1929): 2751–2758.
Miller, Dudley G. Radioactivity and Radiation Detection. Gordon and Breach Science
Publishers, 1972.
97
Miller, George, and Christopher Winstead. “Cavity Ringdown Laser Absorption
Spectroscopy.” In Encyclopedia of Analytical Chemistry, edited by R.A. Meyers,
10734–50. Chichester, England: John Wiley and Sons, Ltd., 2000.
Olive, B.S. “Absorption Spectra of Benzene, Toluene, and Sulfur Dioxide.” Florence,
AL: University of North Alabama, 2015.
———. “Absorption Spectrum of Sulfur Dioxide.” Florence, AL: University of North
Alabama, 2007.
Paldus, Barbara A, and Alexander A Kachanov. “An Historical Overview of CavityEnhanced Methods.” Canadian Journal of Physics 83, no. 10 (October 2005):
975–99. doi:10.1139/p05-054.
Price, William James. Nuclear Radiation Detection. McGraw-Hill, 1964.
Reese, Tyler. “A System for Measuring Radiation Induced Chemical Products in
Atmospheric Gases Using Optical Detection Methods.” Master’s Thesis, The
University of Southern Mississippi, 2010.
Richards, W. G., and P. R. Scott. Structure and Spectra of Molecules. 1 edition.
Chichester West Sussex ; New York: Wiley, 1985.
Roberts–Semple, Dawn, Fei Song, and Yuan Gao. “Seasonal Characteristics of Ambient
Nitrogen Oxides and Ground–level Ozone in Metropolitan Northeastern New
Jersey.” Atmospheric Pollution Research 3, no. 2 (April 2012): 247–57.
doi:10.5094/APR.2012.027.
Rufus, J, G Stark, P.L. Smith, J.C. Pickering, and A.P. Thorne. “High-Resolution
Photoabsorption Cross Section Measurements of SO2, 2: 220 to 325 Nm at 295
K.” J. Geophys. Research - Planets 108 (2003).
98
Sander, S.P. “Temperature Dependence of the NO3 Absorption Spectrum.” J. Phys.
Chem. 90 (1986): 4135–42.
Seinfeld, John H., and Spyros N. Pandis. Atmospheric Chemistry and Physics : From Air
Pollution to Climate Change. A Wiley-Interscience Publication. New York:
Wiley, 1998.
http://lynx.lib.usm.edu/login?url=http://search.ebscohost.com/login.aspx?direct=tr
ue&db=nlebk&AN=26130&site=ehost-live.
Travis, Jeffrey, and Jim Kring. LabVIEW for Everyone: Graphical Programming Made
Easy and Fun. 3rd ed. Upper Saddle River, NJ: Prentice Hall, 2007.
Tsoulfanidis, Nicholas. Measurement and Detection of Radiation. Hemisphere Pub.
Corp., 1983.
Turner, J. E. Atoms, Radiation, and Radiation Protection. 3rd completely rev. and Ed.
Physics Textbook. Weinheim: Wiley-VCH, 2007.
“Understanding Shared Variable Technology - LabVIEW 2011 Help - National
Instruments.” Accessed September 16, 2016. https://zone.ni.com/reference/enXX/help/371361H-01/lvconcepts/ni_psp/.
Vandaele, A.C., and S Fally. “Fourier Transform Measurements of SO2 Absorption
Cross Sections: II. Temperature Dependence in the 29 000-44 000 Cm-1 (227-345
Nm) Region.” J. Quant. Spectrosc. Rad. Transfer 110, no. 18 (2009): 2115–26.
Vandaele, A.C., G Simon, M Carleer, and R Colin. “SO2 Absorption Cross Section
Measurement in the UV Using a Fourier Transform Spectrometer.” J. Geophys.
Res. 99 (1994): 25599–605.
99
Willis, C., A. W. Boyd, and M. J. Young. “Radiolysis of Air and Nitrogen-Oxygen
Mixtures with Intense Electron Pulses: Determination of a Mechanism by
Comparison of Measured and Computed Yields.” Canadian Journal of Chemistry
48, no. 10 (1970): 1515–1525.
100